Managing the information that drives the enterprise Vol. 10 No. 9 November 2011 NAS BUYING TIPS • DR MONITORING APPS STORAGE ALSO INSIDE Get real about the cloud Time is right for SSDs Best fit for scale-out NAS Hybrid clouds gain favor Tale of the tape Storage Pay Still On the Rise With an economy stuck in neutral, storage professionals’ paychecks are still growing— but for job satisfaction, money isn’t everything.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Managing the information that drives the enterprise
Vol 10 No 9 November 2011
NAS BUYING TIPS bull DR MONITORING APPS
STORAGE
ALSO INSIDEGet real about the cloud
Time is right for SSDs
Best fit for scale-out NAS
Hybrid clouds gain favor
Tale of the tape
Storage Pay
StillOn the
RiseWith an economy stuck
in neutral storage professionalsrsquo paychecks
are still growingmdashbut for job satisfaction money isnrsquot everything
FROM OUR SPONSORS
3 STORAGE November 2011
STORAGEinside | november 2011
Letrsquos get real about the cloud5 EDITORIAL Yoursquoll need to look past the ldquoirrational exuberancerdquo of the
cloud storage market to get a real handle on how it might fit into your data storage environment by RICH CASTAGNA
Time is right for SSDs8 RANDOM ACCESS Solid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But new low-cost systems built specifically for solid-state are generating a lot of interestby DAVE RAFFO
Salary survey Storage prosrsquo pay still on the rise11 Even with an economy thatrsquos stubbornly stuck in neutral data storage
professionalsrsquo paychecks reflect modest yet welcome increases by ELLEN OrsquoBRIEN
NAS system buying decisions21 Whether yoursquore dealing with big data issues or just trying to stem the
tide of file data new developments in NAS systems and a wide range of products create a bevy of attractive alternativesby JACOB NORBEL GSOEDL
Disaster recovery readiness monitoring applications33 Developing and implementing disaster recovery plans can be complex and
time-consuming but a new class of apps can help you see if your DR plansare in sync with your IT operations by PAUL KIRVAN CISA FBCI CBCP
Who really needs scale-out systems43 HOT SPOTS Industries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can provide high-performance application supportby TERRI MCCLURE
Cloud storage ahead of the pack for hybrid cloud deployments48 READWRITE The idea of turning over storage systems to the cloud
hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage by JEFF BYRNE
Tape still plays a role in the data center52 SNAPSHOT In our latest Snapshot survey 58 of Storage readers say
theyrsquore using tape as much or more than they did three years ago and only 16 have banished tape entirely by RICH CASTAGNA
From our sponsors54 Useful links from our sponsors
server rooms that require GPs NaviGatioN
We get that virtualization can drive a better ROI Highly certified by Microsoft VMware HP and others we can evaluate design and implement the right solution for youWersquoll get you out of this mess at CDWcomvirtualization
soLveD
copy2011 CDW LLC CDWreg CDWbullGreg and PeOPLe WHO Get Ittrade are trademarks of CDW LLC
c
Storage May 2010Copyright 2011 TechTarget No part of this publication may be transmitted or reproduced in any form or by any means without permission in writingfrom the publisher For permissions or reprint information please contact Mike Kelly VP and Group Publisher (mkellytechtargetcom)
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
5
LOUD STORAGE IS limitless Cloud storage is ubiquitous Cloud storage is elasticCloud storage is economical Cloud storage will bring world peace curecancer and balance the budget
Yes Irsquove bought into all the promises about cloud storage but Irsquom notquite ready to drink the Kool-Aid chant the cloud storage mantra andstagger into the ether like some zombie extra from Night of the Living DeadIrsquod feel a little more comfortable if we could just squeeze the word ldquopo-tentiallyrdquo into those four ldquoCloud storage is rdquo sentences above
Itrsquos not just the relentless vendor hypemdashitrsquos turning into a story aboutexpectation and maybe even just a dab or two of exaggeration
Gartner IDC and just about any analyst firm with at least a toe in thedata storage market waters have pitched predictions of runaway growthfor the cloud computing market The Computing Technology Industry Association (CompTIA) recently issued a report detailing the results of its cloud computing survey In the reportrsquos description of the cloud marketCompTIA cited Gartnerrsquos prediction that ldquocloud storage will grow at 895CAGR to $288 billionrdquo by 2015 The report also noted IDCrsquos prediction whichisnrsquot restricted to just storage ldquoIDC predicts that public cloud IT spendingwill grow from $215 billion in 2010 to $729 billion in 2015rdquo
But CompTIA is careful to point out that current expenditures for cloudservices represent approximately 2 of the whole IT spend so even if itdoubles itrsquos still just a drop in the bucket in the big IT picture
And if Irsquom going to do any finger-pointing I also have to point at us Ourmost recent survey research shows that approximately 21 of companiesuse cloud storage services for non-backup purposes and about 28 ofthem use some form of cloud backup for at least one app Those are prettygood numbers for relatively new technologies but wersquove also seen someups and downs with our stats which might indicate that companies arejust testing cloud storage rather than committing to it
We need a little context to be able to look at cloud storage in a rational
editorial | rich castagna
Letrsquos get real about the cloudYoursquoll need to look past the ldquoirrational exuberancerdquo of the cloud storage market to get a real handle on
how it might fit into your data storage environment
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
FROM OUR SPONSORS
3 STORAGE November 2011
STORAGEinside | november 2011
Letrsquos get real about the cloud5 EDITORIAL Yoursquoll need to look past the ldquoirrational exuberancerdquo of the
cloud storage market to get a real handle on how it might fit into your data storage environment by RICH CASTAGNA
Time is right for SSDs8 RANDOM ACCESS Solid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But new low-cost systems built specifically for solid-state are generating a lot of interestby DAVE RAFFO
Salary survey Storage prosrsquo pay still on the rise11 Even with an economy thatrsquos stubbornly stuck in neutral data storage
professionalsrsquo paychecks reflect modest yet welcome increases by ELLEN OrsquoBRIEN
NAS system buying decisions21 Whether yoursquore dealing with big data issues or just trying to stem the
tide of file data new developments in NAS systems and a wide range of products create a bevy of attractive alternativesby JACOB NORBEL GSOEDL
Disaster recovery readiness monitoring applications33 Developing and implementing disaster recovery plans can be complex and
time-consuming but a new class of apps can help you see if your DR plansare in sync with your IT operations by PAUL KIRVAN CISA FBCI CBCP
Who really needs scale-out systems43 HOT SPOTS Industries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can provide high-performance application supportby TERRI MCCLURE
Cloud storage ahead of the pack for hybrid cloud deployments48 READWRITE The idea of turning over storage systems to the cloud
hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage by JEFF BYRNE
Tape still plays a role in the data center52 SNAPSHOT In our latest Snapshot survey 58 of Storage readers say
theyrsquore using tape as much or more than they did three years ago and only 16 have banished tape entirely by RICH CASTAGNA
From our sponsors54 Useful links from our sponsors
server rooms that require GPs NaviGatioN
We get that virtualization can drive a better ROI Highly certified by Microsoft VMware HP and others we can evaluate design and implement the right solution for youWersquoll get you out of this mess at CDWcomvirtualization
soLveD
copy2011 CDW LLC CDWreg CDWbullGreg and PeOPLe WHO Get Ittrade are trademarks of CDW LLC
c
Storage May 2010Copyright 2011 TechTarget No part of this publication may be transmitted or reproduced in any form or by any means without permission in writingfrom the publisher For permissions or reprint information please contact Mike Kelly VP and Group Publisher (mkellytechtargetcom)
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
5
LOUD STORAGE IS limitless Cloud storage is ubiquitous Cloud storage is elasticCloud storage is economical Cloud storage will bring world peace curecancer and balance the budget
Yes Irsquove bought into all the promises about cloud storage but Irsquom notquite ready to drink the Kool-Aid chant the cloud storage mantra andstagger into the ether like some zombie extra from Night of the Living DeadIrsquod feel a little more comfortable if we could just squeeze the word ldquopo-tentiallyrdquo into those four ldquoCloud storage is rdquo sentences above
Itrsquos not just the relentless vendor hypemdashitrsquos turning into a story aboutexpectation and maybe even just a dab or two of exaggeration
Gartner IDC and just about any analyst firm with at least a toe in thedata storage market waters have pitched predictions of runaway growthfor the cloud computing market The Computing Technology Industry Association (CompTIA) recently issued a report detailing the results of its cloud computing survey In the reportrsquos description of the cloud marketCompTIA cited Gartnerrsquos prediction that ldquocloud storage will grow at 895CAGR to $288 billionrdquo by 2015 The report also noted IDCrsquos prediction whichisnrsquot restricted to just storage ldquoIDC predicts that public cloud IT spendingwill grow from $215 billion in 2010 to $729 billion in 2015rdquo
But CompTIA is careful to point out that current expenditures for cloudservices represent approximately 2 of the whole IT spend so even if itdoubles itrsquos still just a drop in the bucket in the big IT picture
And if Irsquom going to do any finger-pointing I also have to point at us Ourmost recent survey research shows that approximately 21 of companiesuse cloud storage services for non-backup purposes and about 28 ofthem use some form of cloud backup for at least one app Those are prettygood numbers for relatively new technologies but wersquove also seen someups and downs with our stats which might indicate that companies arejust testing cloud storage rather than committing to it
We need a little context to be able to look at cloud storage in a rational
editorial | rich castagna
Letrsquos get real about the cloudYoursquoll need to look past the ldquoirrational exuberancerdquo of the cloud storage market to get a real handle on
how it might fit into your data storage environment
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
3 STORAGE November 2011
STORAGEinside | november 2011
Letrsquos get real about the cloud5 EDITORIAL Yoursquoll need to look past the ldquoirrational exuberancerdquo of the
cloud storage market to get a real handle on how it might fit into your data storage environment by RICH CASTAGNA
Time is right for SSDs8 RANDOM ACCESS Solid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But new low-cost systems built specifically for solid-state are generating a lot of interestby DAVE RAFFO
Salary survey Storage prosrsquo pay still on the rise11 Even with an economy thatrsquos stubbornly stuck in neutral data storage
professionalsrsquo paychecks reflect modest yet welcome increases by ELLEN OrsquoBRIEN
NAS system buying decisions21 Whether yoursquore dealing with big data issues or just trying to stem the
tide of file data new developments in NAS systems and a wide range of products create a bevy of attractive alternativesby JACOB NORBEL GSOEDL
Disaster recovery readiness monitoring applications33 Developing and implementing disaster recovery plans can be complex and
time-consuming but a new class of apps can help you see if your DR plansare in sync with your IT operations by PAUL KIRVAN CISA FBCI CBCP
Who really needs scale-out systems43 HOT SPOTS Industries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can provide high-performance application supportby TERRI MCCLURE
Cloud storage ahead of the pack for hybrid cloud deployments48 READWRITE The idea of turning over storage systems to the cloud
hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage by JEFF BYRNE
Tape still plays a role in the data center52 SNAPSHOT In our latest Snapshot survey 58 of Storage readers say
theyrsquore using tape as much or more than they did three years ago and only 16 have banished tape entirely by RICH CASTAGNA
From our sponsors54 Useful links from our sponsors
server rooms that require GPs NaviGatioN
We get that virtualization can drive a better ROI Highly certified by Microsoft VMware HP and others we can evaluate design and implement the right solution for youWersquoll get you out of this mess at CDWcomvirtualization
soLveD
copy2011 CDW LLC CDWreg CDWbullGreg and PeOPLe WHO Get Ittrade are trademarks of CDW LLC
c
Storage May 2010Copyright 2011 TechTarget No part of this publication may be transmitted or reproduced in any form or by any means without permission in writingfrom the publisher For permissions or reprint information please contact Mike Kelly VP and Group Publisher (mkellytechtargetcom)
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
5
LOUD STORAGE IS limitless Cloud storage is ubiquitous Cloud storage is elasticCloud storage is economical Cloud storage will bring world peace curecancer and balance the budget
Yes Irsquove bought into all the promises about cloud storage but Irsquom notquite ready to drink the Kool-Aid chant the cloud storage mantra andstagger into the ether like some zombie extra from Night of the Living DeadIrsquod feel a little more comfortable if we could just squeeze the word ldquopo-tentiallyrdquo into those four ldquoCloud storage is rdquo sentences above
Itrsquos not just the relentless vendor hypemdashitrsquos turning into a story aboutexpectation and maybe even just a dab or two of exaggeration
Gartner IDC and just about any analyst firm with at least a toe in thedata storage market waters have pitched predictions of runaway growthfor the cloud computing market The Computing Technology Industry Association (CompTIA) recently issued a report detailing the results of its cloud computing survey In the reportrsquos description of the cloud marketCompTIA cited Gartnerrsquos prediction that ldquocloud storage will grow at 895CAGR to $288 billionrdquo by 2015 The report also noted IDCrsquos prediction whichisnrsquot restricted to just storage ldquoIDC predicts that public cloud IT spendingwill grow from $215 billion in 2010 to $729 billion in 2015rdquo
But CompTIA is careful to point out that current expenditures for cloudservices represent approximately 2 of the whole IT spend so even if itdoubles itrsquos still just a drop in the bucket in the big IT picture
And if Irsquom going to do any finger-pointing I also have to point at us Ourmost recent survey research shows that approximately 21 of companiesuse cloud storage services for non-backup purposes and about 28 ofthem use some form of cloud backup for at least one app Those are prettygood numbers for relatively new technologies but wersquove also seen someups and downs with our stats which might indicate that companies arejust testing cloud storage rather than committing to it
We need a little context to be able to look at cloud storage in a rational
editorial | rich castagna
Letrsquos get real about the cloudYoursquoll need to look past the ldquoirrational exuberancerdquo of the cloud storage market to get a real handle on
how it might fit into your data storage environment
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
server rooms that require GPs NaviGatioN
We get that virtualization can drive a better ROI Highly certified by Microsoft VMware HP and others we can evaluate design and implement the right solution for youWersquoll get you out of this mess at CDWcomvirtualization
soLveD
copy2011 CDW LLC CDWreg CDWbullGreg and PeOPLe WHO Get Ittrade are trademarks of CDW LLC
c
Storage May 2010Copyright 2011 TechTarget No part of this publication may be transmitted or reproduced in any form or by any means without permission in writingfrom the publisher For permissions or reprint information please contact Mike Kelly VP and Group Publisher (mkellytechtargetcom)
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
5
LOUD STORAGE IS limitless Cloud storage is ubiquitous Cloud storage is elasticCloud storage is economical Cloud storage will bring world peace curecancer and balance the budget
Yes Irsquove bought into all the promises about cloud storage but Irsquom notquite ready to drink the Kool-Aid chant the cloud storage mantra andstagger into the ether like some zombie extra from Night of the Living DeadIrsquod feel a little more comfortable if we could just squeeze the word ldquopo-tentiallyrdquo into those four ldquoCloud storage is rdquo sentences above
Itrsquos not just the relentless vendor hypemdashitrsquos turning into a story aboutexpectation and maybe even just a dab or two of exaggeration
Gartner IDC and just about any analyst firm with at least a toe in thedata storage market waters have pitched predictions of runaway growthfor the cloud computing market The Computing Technology Industry Association (CompTIA) recently issued a report detailing the results of its cloud computing survey In the reportrsquos description of the cloud marketCompTIA cited Gartnerrsquos prediction that ldquocloud storage will grow at 895CAGR to $288 billionrdquo by 2015 The report also noted IDCrsquos prediction whichisnrsquot restricted to just storage ldquoIDC predicts that public cloud IT spendingwill grow from $215 billion in 2010 to $729 billion in 2015rdquo
But CompTIA is careful to point out that current expenditures for cloudservices represent approximately 2 of the whole IT spend so even if itdoubles itrsquos still just a drop in the bucket in the big IT picture
And if Irsquom going to do any finger-pointing I also have to point at us Ourmost recent survey research shows that approximately 21 of companiesuse cloud storage services for non-backup purposes and about 28 ofthem use some form of cloud backup for at least one app Those are prettygood numbers for relatively new technologies but wersquove also seen someups and downs with our stats which might indicate that companies arejust testing cloud storage rather than committing to it
We need a little context to be able to look at cloud storage in a rational
editorial | rich castagna
Letrsquos get real about the cloudYoursquoll need to look past the ldquoirrational exuberancerdquo of the cloud storage market to get a real handle on
how it might fit into your data storage environment
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
c
Storage May 2010Copyright 2011 TechTarget No part of this publication may be transmitted or reproduced in any form or by any means without permission in writingfrom the publisher For permissions or reprint information please contact Mike Kelly VP and Group Publisher (mkellytechtargetcom)
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
5
LOUD STORAGE IS limitless Cloud storage is ubiquitous Cloud storage is elasticCloud storage is economical Cloud storage will bring world peace curecancer and balance the budget
Yes Irsquove bought into all the promises about cloud storage but Irsquom notquite ready to drink the Kool-Aid chant the cloud storage mantra andstagger into the ether like some zombie extra from Night of the Living DeadIrsquod feel a little more comfortable if we could just squeeze the word ldquopo-tentiallyrdquo into those four ldquoCloud storage is rdquo sentences above
Itrsquos not just the relentless vendor hypemdashitrsquos turning into a story aboutexpectation and maybe even just a dab or two of exaggeration
Gartner IDC and just about any analyst firm with at least a toe in thedata storage market waters have pitched predictions of runaway growthfor the cloud computing market The Computing Technology Industry Association (CompTIA) recently issued a report detailing the results of its cloud computing survey In the reportrsquos description of the cloud marketCompTIA cited Gartnerrsquos prediction that ldquocloud storage will grow at 895CAGR to $288 billionrdquo by 2015 The report also noted IDCrsquos prediction whichisnrsquot restricted to just storage ldquoIDC predicts that public cloud IT spendingwill grow from $215 billion in 2010 to $729 billion in 2015rdquo
But CompTIA is careful to point out that current expenditures for cloudservices represent approximately 2 of the whole IT spend so even if itdoubles itrsquos still just a drop in the bucket in the big IT picture
And if Irsquom going to do any finger-pointing I also have to point at us Ourmost recent survey research shows that approximately 21 of companiesuse cloud storage services for non-backup purposes and about 28 ofthem use some form of cloud backup for at least one app Those are prettygood numbers for relatively new technologies but wersquove also seen someups and downs with our stats which might indicate that companies arejust testing cloud storage rather than committing to it
We need a little context to be able to look at cloud storage in a rational
editorial | rich castagna
Letrsquos get real about the cloudYoursquoll need to look past the ldquoirrational exuberancerdquo of the cloud storage market to get a real handle on
how it might fit into your data storage environment
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
6 STORAGE November 2011
manner and to see where it may be appropriate as an additionreplacementto on-premises storage Billion-dollar predictions and multi-digit growthrates arenrsquot nearly as impressive if you consider the starting point If Irsquomonly spending $1 today but plan to spend $2 next year itrsquos a 100 hike butitrsquos still just two bucks
Approaching cloud storage from a practical point of view will yield muchbetter results for users and providers alike ldquoCloud-based email storagebackup and recovery and business productivity applications are in the high-est demand among customers buying cloud solutions todayrdquo the CompTIAreport noted And itrsquos hardly surprising these are the most mature areas ofcloud storage and probably even cloud computing as a whole
On the horizon I expect mobilecomputing to be a bigmdashand maybethe mainmdashincentive for companiesto buy into cloud storage servicesSmartphones tablets and whateverultraportable form factor is aroundthe corner pose special problemsfor IT and storage managers inparticular These devices can createa situation in which corporate datais created modified and deletedwithout it ever being stored ondata center gear or maybe without it ever even passing through the datacenter You could argue that the mobile scenario already exists with remoteoffices but remote offices have a physical presence that could be tiedrelatively easily into centralized resources if desired
Although the vendors tend to paint an all-or-nothing picture of cloudstorage (ldquoNo more backupsrdquo) itrsquos likely that hybrid solutions that integratecloud resources with installed systems will best satisfy our needs forsome time to come Itrsquos a reasonable approach thatrsquos relatively free fromhyperbole but there are still just a handful of players trying to bridgethose two environments I expect this market will be ldquojustifiedrdquo as soon as both the heavy-metal and cloud guys in the storage industry realizethat working together just makes sense 2
Rich Castagna (rcastagnastoragemagazinecom) is editorial director of the Storage Media Group
Click here for a sneak peek at whatrsquos coming up in the December 2011 issue
On the horizon I expect mobile computing to be abigmdashand maybe themainmdashincentive forcompanies to buy intocloud storage services
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
EMC PRESENTS
EMC2 EMC and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries All other trademarks are the property of their respective ownerscopy Copyright 2011 EMC Corporation All rights reserved Source IDC Worldwide Purpose Built Backup Appliance 2010-2015 Market Analysis and Forecast 2010 Vendor Shares report (May 2011)
ldquoDiscover the Power of Next Generation Backuprdquo
See why EMC is the leader in backup and recovery at wwwEMCcomtransformbackup
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
8 STORAGE November 2011
nO MATTER HOW much hype it receives no storage technology is welcomedwith open arms in a data center It always has to kick the door down
Even technologies that have become established such as iSCSI anddata deduplication had periods of doubt and mistrust before they scoredbig Large vendors are often slow to adopt the new technologies andusers arenrsquot sure what to make of them But vendors such as EqualLogicand LeftHand showed that iSCSI could make life easier on administratorsand maybe save them some money in the process Data Domain did thesame for data deduplication and backup
Now we could be at a similar breakthrough point with solid-state storageSolid-state isnrsquot a new technology but it was a latecomer to enterprisestorage When it arrived a few years ago it promised no cost savings but a performance boost It was offered first by EMC and soon after by the other large storage vendors But their attempts to fit solid-state drives(SSDs) into storage arrays built for spinning disk negated much of the performance gain and its high cost remained a stumbling block
ldquoFlash was supposed to be revolutionary in the data center but mostpeople couldnrsquot afford the revolutionrdquo said Scott Dietzen CEO at flash SAN startup Pure Storage
Pure Storage is among a new generation of vendors with SSD systemsthat are promising another revolution similar to the ones that spawnediSCSI and dedupe Nimbus Data Systems started it in 2010 and others suchas Kaminario Pure Storage and SolidFire have followed with all-flash SANsXIO (the storage company formerly known as Xiotech) has adapted itsunique ISE architecture into a hybrid system with flash and spinning disk
These systems were built or modified specifically for solid-state andcosts are kept down by including management features that the estab-lished disk vendors usually charge extra for Theyrsquore not only about SSDthey include features storage administrators have come to rely on such
random access | dave raffo
Time is right for SSDsSolid-state was a latecomer to enterprise storage
and had to be retrofitted into data centers But a newgeneration of systems built specifically for solid-state
are keeping costs down and interesting users
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
as snapshots load balancing dedupe high availability and thin provisioningFactoring in dedupe some of these vendors say they can sell you all-
flash arrays for close to $10 a gigabyte even less in some cases If theycan do that and make their systems reliable and efficient that will be anaffordable revolution
Itrsquos too early to say if any of thesevendors will be successful but eBayhas already installed more than 100TB of Nimbus storage If these newSSD products make it big you can expect to see the likes of Dell EMCHewlett-Packard Hitachi Data Sys-tems IBM and NetApp follow in thefootsteps of upstart vendors The established vendors have made somemovement Most now offer lowercost multi-level cell (MLC) flash drives with software or firmware that improves the reliability of MLC drives NetApp broke from the herd by usingflash as cache in its arrays and EMC is preparing systems with flash-basedPCIe cards in servers EMC also sells all-flash storage systems but theyrsquorebasic VNX and VMAX arrays with all SSDs and controllers built for hard drives
The next step would be for the big vendors to develop controllers forSSDs instead of retrofitting disk controllers
For years disk vendors have talked about how they would replace tapeAnd they might soon make good on the threat disk may replace tape atthe top of the media endangered species list But these new flash vendorsare talking about sending spinning disk to the technology museum just as disk vendors claimed they would do to tape
Of course spinning disk will never go away completely For all the gainsit has made as a backup medium disk still hasnrsquot completely replaced tapejust as open systems havenrsquot killed the mainframe and Ethernet will neverkill Fibre Channel in storage networking I also hear stories of people stillusing physical (non-virtual) servers although I canrsquot confirm those rumorsby reading this or other IT publications
But solid-statersquos time is coming in storage and it will reduce the relianceon spinning disk especially as the top storage tier Who knows maybe therewill be a place for hard drives as an archive tier 2
Dave Raffo is senior news director for TechTargetrsquos Storage Media Group publications
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
9 STORAGE November 2011
Itrsquos too early to say ifany of these vendorswill be successfulbut eBay has alreadyinstalled more than100 TB of Nimbusstorage
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Quantumrsquos DXi-Series Appliances with deduplication provide higher performance at lower cost than the leading competitor
Preserving The Worldrsquos Most Important Data Yourstrade
Contact us to learn more at (866) 809-5230 or visit wwwquantumcomdxi
copy2011 Quantum Corporation All rights reserved
Quantum has helped some of the largest organizations in the world integrate
deduplication into their backup process The benefi ts they report are immediate and
signifi cantmdashfaster backup and restore 90+ reduction in disk needs automated DR
using remote replication reduced administration timemdashall while lowering overall costs
and improving the bottom line
Our award-winning DXireg-Series appliances deliver a smart time-saving approach
to disk backup They are acknowledged technical leaders In fact our DXi6500 was
just nominated as a ldquoBest Backup Hardwarerdquo fi nalist in Storage Magazinersquos Best
Product of the Year Awardsmdashitrsquos both faster and up to 45 less expensive than the
leading competitor
Get more bang for your backup today
Faster performance Easier deployment Lower cost
provide higher pleading competi
Preserving The Worldrsquos Most Importan
Contact us to learn more at (8
copy2011 Quantum Corporation All rights reserved
Q
d
s
u
a
O
t
j
P
le
G
F
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE November 2011
Storage prosrsquo pay
still on the rise
Even with an economy thatrsquos stubbornly stuck in neutral data storage professionalsrsquo
paychecks reflect modest yet welcome increases
BY ELLEN OrsquoBRIEN
DATA STORAGE PROFESSIONALS are holding their own during precariouseconomic times according to the ninth annual Storage magazineSearchStoragecom Salary Survey Theyrsquore managing to increase theiroverall average salaries and stay optimistic that their paychecks willcontinue to grow in 2012
Despite ongoing unemployment issues and fears of a double-dip recession storage pros who responded to our in-depth survey regardingtheir salaries careers incentives budgets and benefits managed to increase their year-over-year earnings for the ninth year running
When it came to storage projects our 266 survey respondents seemto fall almost evenly on either side of the recession vs recovery argu-ment approximately half reported they were in ldquomaintenance moderdquowith budgets restricted The other half detailed ambitious projects suchas multisite data center migrations creating virtual environments andmoving to private clouds
Asked to become experts in more technologies each year this yearrsquoscrop of respondents expressed a shared concern there isnrsquot enoughtime in the day to learn everything there is to know about todayrsquos datastorage market
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
11
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
12 STORAGE November 2011
Our 2011 Salary Survey respondents earned an average of $86926 a
5 jump from their 2010salaries When lookingahead to 2012 this yearrsquosrespondents predictedtheir average annualsalaries would jump by another 4 to $90392
This yearrsquos respondentssaw a larger year-over-year increase than lastyearrsquos crop who barelymanaged a 1 increase inpay vs 2009 Participantswere also a bit gloomierlast year predicting theirsalaries wouldnrsquot budge in the coming year andmight even drop The current grouprsquos estimated4 jump would translateto nearly a 10 raise between 2010 and 2012
BIG PICTURE
Our 2011 respondents reported salariesthat were 5 higher than last year
DRILL DOWN
This yearrsquos crop of respondents saw agreater year-over-year salary increase thanlast yearrsquos respondents They also beat outthe 2009 class of survey participants whenit came to annual increase in pay
QUOTABLE
ldquoEven with the economy the way it is good resources are being kept within organizations and if you are good at what you do companies are always looking for good resourcesrdquo
$0
20000
40000
60000
80000
100000
201220112010
$90392$86926$82668
Average Salary Looking Back and Looking Ahead
Salaries stretch outlook brightens
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Respondents expresseda high level of satis-faction with their
employee benefits in the four categories we surveyed health dental vacation and flex-time A little more than 60 rated their health benefitsas good very good or excellent
The majority of respon-dents (679) said theirbenefits packages didnrsquotchange compared to lastyear 10 cited improvedbenefits while 196 sawthem reducedmdashnot a veryencouraging number but adefinite improvement vs the30 who fell into that category last year
BIG PICTURE
Overall satisfaction with company benefitsplans was high
DRILL DOWN
196 saw benefits reduced and some reported being asked for higher co-pays
QUOTABLE
ldquoIrsquom being forced to take time [off] insteadof getting overtime payrdquo
My company does not provide a benefits package
My benefits package was improved
My benefits package was reduced
My benefits did not change
679196
1025
Review of Benefits Package2011 vs 2010
Benefits scorecard Some cuts and high marks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
13 STORAGE November 2011
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
In 2011 storage prosmanaging the largestdata storesmdashmore
than 500 TBmdashcommandedthe highest annual averagesalary at $110085
Otherwise the resultswerenrsquot conclusive enoughto say that paychecksgrew right alongside theamount of data managed
There was also a steady rise in compensation as it related to the number ofdirect reports Storage prosmanaging small teamswith fewer than five peopleearned $81608 and the average salary climbedsteadily to $122667 forthose managing between21 and 50 IT team members
BIG PICTURE
Storage pros managing the largest teamsand the most terabytes earned the highestsalaries
DRILL DOWN
Just as with terabytes bigger budgets addup to higher pay Managers with budgetsless than $500000 earned average annualsalaries of $77140 Those managing budgetsbetween $11 million and $5 million earned$100740
QUOTABLE
ldquoMy ideal job would be minemdashbut withmore skilled staff around merdquo
$0 20000 40000 60000 80000 100000 120000
More than 500 TB
100 TB to 500 TB
10 TB to 99 TB
1 TB to 9 TB
Less than 1 TB
None $86818
$59500
$78938
$84384
$78180
$110085
Average 2011 Salary Based onTerabytes Managed
How many terabytes are in your paycheck
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
14 STORAGE November 2011
STORAGE
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Storage designed for virtualization keeps custom golf club maker on par with progress
Visit wwwecientvirtualstoragecom to learn how your organization can drive greater
eciency and flexibility with Fluid Data storage from Dell
Scan the QR code with your smart phone to see how Fluid Data gives PING the freedom to innovate and grow as a business or visit wwwcompellentcomPINGDrivesInnovation
PING drives innovation with Fluid Datatrade storage
Eric Hart NetworkInfrastructure Manager PING
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Respondents from thePacific region (AlaskaCalifornia Hawaii
Oregon and Washington)averaged salaries of$95526 pushing them tofirst place after the regionranked fourth in last yearrsquossurvey New England and the Northwest andSouthwest regions werebunched closely finishingsecond third and fourthrespectively
The Southeast regionbrought up the rear (as it did last year) with an average annual salary of$79161 Slightly higher was the Mountain region(repeating last yearrsquos placement) with an aver-age salary of $79837
In Canada data storageprofessionals reportedearning an average annualsalary of $85938 puttingthem ahead of two US regions
BIG PICTURE
The Pacific region took top honors in theregional salary comparison Last yearrsquoswinner New England ranked second
DRILL DOWN
Nearly all our regional divisions rankedclosely together in terms of salary onlythe Mountain and Southeast regions averaged below $80000
QUOTABLE
ldquoPhiladelphia is a small IT community Ihave made good connections with people I see again and againrdquo
$0 20000 40000 60000 80000 100000
Southeast
Mountain
Canada
Midwest
Mid-Atlantic
Southwest
Northwest
New England
Pacific $95526
$92750
$91667
$91475
$90364
$87655
$85938
$79837
$79161
Average 2011 Salary by Region
Pacific region wins by a nose
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
16 STORAGE November 2011
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
College graduatesearned more than highschool graduates but
only by a sliver Howeveryears of dedicated datastorage experience defi-nitely make a differencemdashno matter what the level of education A collegegraduate with less thantwo years of dedicatedstorage experience aver-aged $78064 With sixyears to 10 years of dedi-cated storage experiencesalaries rose to $88303
A warning to newbies In tough economic timesmany companies weed out applicants withoutdiplomas
Another way to increasesalaries was to have onetwo or three certificationswith each one boostingearning potential accord-ing to our survey Afterthree however the mathstopped working and paychecks leveled off
BIG PICTURE
Years on the job count but the real payoff is in years of dedicated storage experience
DRILL DOWN
College graduates earn about the same asIT pros who attended college for two yearsThe highest paid group of survey respon-dents held Masterrsquos degrees and had morethan 10 years of dedicated storage experience
QUOTABLE
ldquoIrsquom always learning something new Itrsquoschallengingrdquo
$0 20000 40000 60000 80000 100000
More than 20 years
16 to 20 years
11 to 15 years
Six to 10 years
One to five years $65714
$68561
$81621
$95049
$94772
Average 2011 Salary as It Relates to Years in IT
In a complex market experience counts
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
17 STORAGE November 2011
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
A little over 65 of respondents citedsalary as the No 1
factor in their choice ofjob Career advancementwas the second reason cited by respondents forchanging or taking a new job
When asked about theleast important factors intheir job choice benefits(433) and location (254)landed at the bottom of thepriority list
Only 13 of respondentsvoted salary as their leastimportant factor
BIG PICTURE
When choosing a job salary trumps allother considerations
DRILL DOWN
Career advancement ranks second in considerations while location and benefitscompeted for least important factors
QUOTABLE
ldquoMy salary is representative of a commitment the company has to ITrdquo
Other
Benefits
LocationJob responsibilities
Career advancement
Salary
OtherSalary
Career advancementJob responsibilities
Location
Benefits
654158
6767
21 33
433
254
133
133
13 34
Most Important Factors in Job Choice
Least Important Factors in Job Choice
Salary career advancement are chief motivators
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
18 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
19 STORAGE November 2011
In 2011 survey partici-pants were squarelyfocused on that most
valuable of resourcestime Whether it was thespeed at which projectsget completed the hoursspent on documentation or red tapemdashor too manyhours spent in a car andtoo few at the dinnertablemdashrespondents weremade happy or unhappydepending on how muchcontrol they had over theirtime
Long hours long com-mutes and projects withelusive end dates were allreasons for job dissatisfac-tion Working at home thesatisfaction in meetingdeadlines and seeing results and the ability towork fast without a lot ofhigher ups having to signoff on every move were all named as reasons toappreciate their currentpositions 2
Ellen OrsquoBrien is the Executive Editor for theSearchStoragecom andSearchCloudStoragecom websites
BIG PICTURE
Smart peers and respect from the businessside of the company go a long way
DRILL DOWN
Business leaders who undervalue storageand budget cuts continue to plague storagepros Work-life balance helps ease the pain
QUOTABLE
ldquoItrsquos a great place to work Itrsquos not oftenthat I miss the kidsrsquo soccer gamesrdquo
Three reasons to
ldquoThe diversity of the technology I get to work withrdquo
ldquoBeing allowed to move on with [projects] 90 of the timerdquoldquoRelaxed environment and
lots of flexibilityrdquo
ldquoNo official work-from-home policyrdquoldquoLack of communication between
levels of managementrdquoldquoDealing with people that have unreasonable expectationsrdquo
rave
Three reasons to rant
The time of their lives
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
IMPROVEDUTILIZATION
CENTRALIZEDMANAGEMENT
RED
UC
EC
OM
PLE
XIT
Y
VO
LUM
EM
AN
AG
EMEN
T
STORAGEMANAGEMENT
CO
NTR
OL
VIR
TUA
LEN
VIR
ON
MEN
TS
MA
XIM
IZE
STO
RA
GE
CA
PAC
ITY
REDUCEDOPERATIONAL COSTS
INTELLIGENTARCHIVING
DATA PROTECTION
STORAGEOPTIMIZATION
END-TO-ENDVISIBILITY H
IGH
AVA
ILA
BIL
ITY
CEN
TRA
LLY
MA
NA
GED
INC
RE
AS
E D
ATA
AV
AIL
AB
ILIT
Y
RED
UC
EDR
ISK
SHETEROGENEOUSOPERATING SYSTEMS
CAPACITYMANAGEMENTP
ERFO
RM
AN
CE
MIG
RA
TIO
N
CLU
STE
RIN
G
CONSOLIDATESERVERS
CO
ST
EFFE
CTI
VE
MITIGATEDOWNTIME
GREENEFFICIENCY
PO
WER
SA
VIN
GS
DATACENTEREFFICIENCY
MAXIMIZE DATASTORAGE CAPACITY
MAXIMIZEDATA STORAGE
REDUCE DATA VOLUMESREDUCE BACKUP DATA
DATACENTERSTORAGE
AR
CH
IVE
UN
STR
UC
TUR
ED
DA
TA
POLICYBASEDARCHIVING
IDEN
TIFY
UN
US
EDS
TOR
AG
EC
APA
CIT
Y
IDENTIFY UNUSEDDATA CENTER STORAGE
EFFICIENT DATACENTER STORAGE
RECLAIM
LOSTDATA CENTERSTORAGE
RECLAIMDATA CENTERSTORAGE S
TOR
AG
E EF
FIC
IEN
CY
STORAGE SOFTWARE
SYMANTEC IS
THE LEADER IN99 OF THE FORTUNE 500reg RELY ON SYMANTEC STORAGE SOLUTIONSSecure optimize and manage data more efficiently at gosymanteccomstoragesoftwareleader
copy 2010 Symantec Corporation All rights reserved Symantec and the Symantec logo are registered trademarks of Symantec Corporation or its affiliates in the US and other countries Other names may be trademarks of their respective owners
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
eGet real about
the cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
21 STORAGE November 2011
XPONENTIAL GROWTH OF unstructured data continues to spur file storagegrowth and network-attached storage (NAS) deployments According to a research study done by Milford Mass-based Enterprise Strategy Group(ESG) one-fifth of users report NAS capacity growth of more than 50 per year and 54 of organizations with NAS installations said their storagecapacity is growing by at least 20 per year Most companies opt for NASstorage rather than file shares on servers because of the reliability perform-
NAS SYSTEM BUYING
DECISIONSWhether yoursquore dealing with big data issues or just tryingto stem the tide of file data new developments in NASsystems and a wide range of products put them centerstage as attractive alternatives BY JACOB NORBEL GSOEDL
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
ance scalability and storage management advantages of NAS as well asfeatures such as replication snapshotting thin provisioning and efficientcloning Available as gateways to front-end block-based storage and ascomplete systems with their own storage NAS systems are now used by companies ranging from very small outfits to large enterprises
NAS MARKET SEGMENTATIONAt the low end of the NAS spectrum are systems for home users Priced aslow as $100 they may come with two or four drives with an option for RAID01 and include features such as a single gigabit Ethernet port web-basedadministration the ability to create users and file shares support for basicquotas for file shares and a backup application With modest performanceand no ability to scale theyrsquore intended for ldquoprosumerrdquo users and the smallofficehome office (SOHO) market Examples of these entry-level NAS sys-tems are D-Link ShareCenter EMCrsquos Iomega StorCenter Netgear Stora andSeagate BlackArmor families
Low-end to midrange NAS appliances that are usually found in smaller organizations are the next level up Capacities can range from a few terabytesto more than 100 TB with some supporting SAS in addition to SATA drivesThese systems usually offer some high-end protocols and features such as various RAID levels Microsoft Active Directory (AD) support snapshotsreplication and a dedicated backup interface Appropriately sized with multicore CPUs sufficient memory SAS drive options and multiple gigabitports they deliver decent performance Usually priced under $25000theyrsquore intended for small offices and remote offices of larger organizationswith limited IT staffs and budgets Usability and support options are key requirements in this segment The various Microsoft Windows StorageServer-based NAS offerings Netgear ReadyNAS and Nexsan E5000 familiesand Overland Storage SnapServer family are examples of systems in thisNAS segment At the higher end of the spectrum NetApp with its FAS2000family is also targeting this space
ldquoThe focus of low-end to midrange NAS systems is to combine ease ofuse with affordability and pertinent NAS features such as AD integrationreplication and snapshotsrdquo said Greg Schulz founder and senior analyst at Stillwater Minn-based StorageIO Group
Moving farther up the ladder are NAS systems for small- and medium-sized businesses (SMBs) and enterprises This segment has been dominatedby EMC and NetApp with BlueArc (now part of Hitachi Data Systems) and
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
22 STORAGE November 2011
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
23 STORAGE November 2011
HomeSOHO NAS
Modest
Not available
RAID 1 andsometimesRAID 5
Very limited
Very low
MidrangeNAS
Adequate
Very limited
Critical componentsare redun-dant dual-controlleroptions for highavailability
Provides essentialfeatures
Very affordable
Enterprisescale-upNAS
Usually opti-mized for ahigh numberof IOs
Scales vertically by addingspindles
No singlepoints of failure dual-controller architecturefor highavailability
Most completefeature set
High
Scale-outNAS
High numberof IOs and highthroughput
Very goodproportion-ally scalesperformanceand capacityby addingnodes
No singlepoints of failure resilient to multiplefailures at one time
Lagging enterprisescale-upNAS systems
High
Performance
Scalability
Redundancy
Features
Cost
NAS system segments
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
emerging scale-out NAS vendors on their heels ldquoIn enterprise deals you always see NetApp or EMC or both and sometimes some of the other playersrdquo Schulz said EMCrsquos acquisition of Isilon Systems and NetApprsquos support of scale-out deployments in Data Ontap 8 help tilt the odds towardEMC and NetApp
High performance support for hun-dreds of terabytes up to petabytes of capacity high availability (HA) enterprise-level support and an ever-growing list of NAS features arecharacteristic of this product classEnterprise NAS systems are pushingthe feature envelope in a seeminglycontinuous attempt to expand featurerequirements For instance supportfor both block- and file-based proto-cols the ability to scale horizontally and deduplication are changing fromnice-to-haves to must-haves Enterprise NAS systems are usually optimizedfor traditional corporate apps that is to excel in serving a large number ofsmall files and to display good performance with back-office apps such asMicrosoft SharePoint Exchange and SQL Server but theyrsquore usually not thefirst choice for applications that push throughput limits
Certain applications and industriesmdashsuch as the mediaentertainmentindustry and gasoil exploration where very large files rather than a lot ofsmaller files prevailmdashrequire very high throughput and massive scalabilityThis has been the sweet spot of scale-out and high-performance NASsystems such as EMC Isilon Hewlett-Packard (HP) StorageWorks X9000Network Storage Systems IBM Scale Out Network Attached Storage(SONAS) and Panasas
NAS FEATURES IN DEPTHRegardless of your company size or file-storage needs itrsquos essential toconsider the key NAS attributes and features to ensure yoursquoll arrive at abuying decision you wonrsquot regret later These are the key areas to look atwhen evaluating NAS systems
bull Dual controller vs scale-out NAS architecturebull Storage efficiencybull Unified storagebull Virtualization
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
24 STORAGE November 2011
ldquoIn enterprise dealsyou always seeNetApp or EMC orboth and sometimessome of the otherplayersrdquo
mdashGREG SCHULZ founder and senior analyst StorageIO Group
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
DUAL CONTROLLER VS SCALE-OUT NAS ARCHITECTUREFully redundant dual-controller NAS systems have dominated the enterprisespace since they first appeared on the market in the early 1990s Theyscale vertically by adding disk spindles and all components are tightly coupled and share a common pool of resources Once the performancelimit is reached the only scaling option is to add additional NAS systemsthat operate independently of each other Because they scale by addingspindles they perform well with workloads that primarily consist of randomaccess of small files which represents the prevailing workload in enterprisedata centers
Conversely scale-out NAS consistsof loosely coupled processing nodesstarting with as few as two that oper-ate in parallel and scale horizontally byadding additional nodes Even thoughthe degree of parallelism amongscale-out systems varies in generalprocessing nodes work in concert todeliver files Each time a node is addedperformance of the overall system increases proportionally For the mostpart a scale-out NAS approach elimi-nates the need for expensive forkliftupgrades to replace existing systemsinstead it makes it possible for the existing single NAS system to expandto multiple petabytes The benefits ofscale-out NAS are pretty compelling improved performance of both IO andthroughput improved scalability cost reduction simplified managementby only having to manage a single large system and elevated high avail-ability ldquoWe predict that by 2015 80 of all storage systems [NAS and blockbased] will be based on scale-out designsrdquo said Terri McClure a senior analyst at ESG
The ability to support scale-out deployments is increasingly becoming arequirement for enterprise NAS systems and traditional scale-up systemswill be gradually relegated to the SMB space But even in the SMB spacetheyrsquoll face the scale-out threat especially from NetApp with its scale-outdeployment support in Data Ontap 8 Perhaps the strongest indicator wersquovereached a scale-out inflection point is that the leading NAS vendors have
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
25 STORAGE November 2011
For the most part a scale-out NASapproach eliminatesthe need for expen-sive forklift upgradesto replace existingsystems instead it makes it possiblefor the existing single NAS systemto expand to multiplepetabytes
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Your Information is at RiskProtect What Matters Most
As the amount of information your organization has to manage and protect continues to grow the challenge
of managing the potential risk increases exponentially How can you ensure your organizationrsquos information
is not at risk Partner with the company thousands have trusted to store protect and manage their
information regardless of format mdash Iron Mountain With unmatched experience putting us at your side makes
information easier to manage We can do more together
Safeguard your Information Visit us at ironmountaincom
categoRIze aRcHIVe IMage dIScoVeR deStRoY
copy2011 Iron Mountain Incorporated All rights reserved Iron Mountain and the design of the mountain are registered trademarks of Iron Mountain Incorporated in the US and other countries
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
all signed up to the scale-out paradigm Dell through its acquisition of Exanet and having released initial gateway products based on Exanet (theNX3500 a NAS gateway for the PowerVault iSCSI arrays and the NX7500 a NAS gateway for the EqualLogic array family) EMC by acquiring Isilon HP with StorageWorks X9000 IBM with SONAS and NetApp
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
27 STORAGE November 2011
The state of NFStTHE NETWORK FILE SYSTEM (NFS) protocol was originally developed bySun Microsystems and quickly evolved to the standard on Unix- andLinux-based operating systems Supported by most NAS systems in addition to Server Message BlockCommon Internet File System(SMBCIFS used with Microsoft Windows systems) itrsquos becomingeven more relevant as a result of sev-eral recent developments CurrentlyNFS Version 3 is prevalent and used bymost companies ldquoVery few of our cus-tomers use NFS Version 4 at this pointrdquosaid Drew Schlussel EMCrsquos director of product management for VNX
NFSv4 has been available for severalyears and stands out for its advancedsecurity features and better perform-ance NFSv41 which was ratified in2010 supports clustered systems andthe ability to provide parallel access to files distributed across multiplenodes via the parallel NFS (pNFS) extension As scale-out NAS architec-tures move into the mainstream NFSv41rsquos support of pNFS will likelyspur adoption VMwarersquos NFS storage support and enhancementsare further popularizing NFS ldquoWersquore seeing NFS becoming morepopular for VMware rather than going with block-based storage[iSCSIFibre Channel]rdquo Schlussel said
mdashDREW SCHLUSSEL director ofproduct management for VNX EMC
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE EFFICIENCYFeatures and capabilities that allow you to store more data on physicaldisks should be a central piece of any storage array evaluation The costsavings of having to buy less physical storage can be significant and caneasily offset higher acquisition costs In other words a more expensiveNAS may cost you less over its lifetime than a less-expensive one with inferior storage efficiency features such as thin provisioning efficientsnapshots thin clones deduplication and compression Implementationand efficacy of these features vary among NAS products One area of differentiation is the granularity at which these features operate In themajority of NAS systems theyrsquore applied to volumes but in some they canbe applied to directories or even files ldquoThe EMC Isilon supports snapshotsreplication and quotas at a file and directory levelrdquo said Sam Grocott vicepresident of marketing at EMC Isilon
The following NAS storage efficiency features should be considered tominimize physical storage requirements
Thin provisioning The ability to allocate more storage than is physicallyavailable is pertinent to achieving high storage utilization This is especiallyimportant in systems that support both file- and block-based protocolswhere thin provisioning enables volumes and NAS pools to be sized inde-pendently of actual physical storage and where physical storage is assignedfrom a common storage pool on an as-needed basis Without thin provi-sioning sufficient physical storage has to be allocated to each volume and storage pool ahead of time In systems that support thin provisioningphysical storage is allocated dynamically when needed
Efficient snapshots Snapshots in NAS systems are invaluable for dataprotection Theyrsquore scheduled to be taken periodically and can optionally be replicated to other NAS systems for disaster recovery (DR) or other dataprotection purposes Supported by most contemporary NAS systems effi-cient snapshots copy data changes and use a system of pointers to refer-ence the initial full snapshot Efficient snapshots not only save valuabledisk space but also reduce the time it takes to complete them minimizingthe performance impact while snapshots are taken
Thin clones Relevant in NAS systems that support block-based protocolsthin clones require little to no storage on creation Similar to efficient snap-shots thinly cloned volumes refer to the original volume via pointers Onlydata on the cloned volume that changes needs to be stored Pioneered byNetApp with FlexClone thin clones are now supported by a growing list ofNAS vendors
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
28 STORAGE November 2011
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Deduplication and compression Standard in backup and archival prod-ucts data deduplication and compression are becoming more prevalent onprimary storage systems While support for deduplication in NAS storage isstill sparse it can be implemented as a post-process scheduled task or inreal-time For instance deduplication in NetApp filers can be enabled on aper-volume basis and is performed by a scheduled process that deduplicates4 KB blocks of data usually during off hours On the contrary the OracleSun ZFS Storage 7000 series performs deduplication in real-time while datais written to disk
Automated storage tiering The ability to keep active data on fast expensive storage and move inactive data to less-expensive slower tiershelps to limit the amount of expensive tier-1 storage neededwithout significantly impactingperformance For any NAS systemyou consider data movement between the different tiers (solid-state storage fast SAS tiers andslower high-capacity SATA tiers)should be automatic with block- or byte-level granularity ratherthan the volume-level granularityat which data is moved The moregranular the better Some systemslike EMCrsquos Fully Automated StorageTiering (FAST) depend on policiesthat define when data should bemoved others like NetApp and Oracle (in the Sun ZFS Storage 7000 series)advocate that the storage system should be smart enough to automatical-ly keep data on the appropriate tier without requiring user-defined policies
UNIFIED STORAGEWith the two leading NAS vendorsmdashEMC in its VNX array family and NetAppin all its systemsmdashsupporting file-system protocols (NFS and CIFS) andblock-based protocols (Fibre Channel and iSCSI) in a single storage arraythe unified storage approach is gaining popularity This is especially true for SMB companies looking for a single storage system that meets all their
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
29 STORAGE November 2011
The ability to keepactive data on fastexpensive storage andmove inactive data toless-expensive slowertiers helps to limit theamount of expensivetier-1 storage neededwithout significantlyimpacting performance
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
storage needs deliver file shares and services but also serve as storage for virtualized servers and back-office applications such as Microsoft SQLServer SharePoint and databases
ldquoSmall and midsize companies are usually looking for a jack-of-all-trades storage system thatrsquos easy tomanage and affordablerdquo said DrewSchlussel EMCrsquos director of productmanagement for VNX
STORAGE AND VIRTUALIZATIONWith the accelerated deployment ofvirtualized servers integration withand features that aid virtualized com-puting infrastructures are becomingimportant storage evaluation criteria It starts with determining how quicklyand efficiently volumes can be assigned to virtual machines (VM) ldquoWithFlexClone we can rapidly clone VMware VMDK images with the click of abuttonrdquo said Jason Blosil NetApprsquos product marketing manager
The simplicity of deploying virtualized servers whose VM images arestored on a single NAS and if created as thin clones share many of thesame physical storage blocks can easily result in performance problemsIdentifying the mechanisms a NAS system has in place to support theunique requirements of virtualized servers is crucial to preventing un-pleasant performance surprises as the number of VMs increases Somevendors use a combination of solid-state storage and automated storagetiering via policies or cache to keep hotspots in the fastest storage tier
VMware has been providing the vStorage API for Array Integration (VAAI)to offload storage and associated management functions from vSphere to storage systems but prior to vSphere 5 support was mostly limited toblock-based storage In vSphere 5 VAAI has been enhanced to better sup-port NFS and network-attached storage for tasks like thin provisioning andsnapshots VAAI support helps remove storage bottlenecks by offloadingresources of intense storage and management tasks from the hypervisorto the storage system
Finally hypervisor plug-ins for VMware vCenter and Microsoft SystemCenter Virtual Machine Manager (SCVMM) that provide storage management
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
30 STORAGE November 2011
ldquoSmall and midsizecompanies are usually looking for a jack-of-all-tradesstorage systemthatrsquos easy to man-age and affordablerdquo
mdashDREW SCHLUSSEL director of product management for VNX EMC
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
and reporting from within hypervisor consoles enable IT generalists andserver administrators to perform storage tasks
EVOLVING NAS SYSTEMSBecause of their many merits NAS systems in the form of gateways toblock-based storage and standalone systems with their own storage havebecome an essential component of todayrsquos data centers With increasedsupport for scale-out architectures block-based protocols along with file-system protocols tight integration with virtualized environments and storageclouds and an expanding list of features NAS system are moving closer tobecoming a storage system nirvana a single fast and massively scalablestorage system for all applications and data 2
Jacob Norbel Gsoedl is a freelance writer and a corporate director for businesssystems He can be reached at jgsoedlyahoocom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
31 STORAGE November 2011
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Up to 85 of computing capacity sits idle in distributed environmentsA smarter planet needs smarter infrastructureLetrsquos build a smarter planet ibmcomdynamic
IBM the IBM logo and ibmcom are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide A current list of IBM trademarks is available on the Web at ldquoCopyright and trademark informationrdquo at wwwibmcomlegalcopytradeshtml
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
33 STORAGE November 2011
DISASTER RECOVERY (DR) planning is typically a significant undertaking that requiresmany work hours and often a considerable budget But creating a plan is only part ofthe process the plan must be tested frequently enough to ensure it will work as expected But testing is a time-consuming and often disruptive activity so itrsquos often notdone as regularly as it should be A new category of applicationsmdashDR readinessmonitoring appsmdashcan facilitate the testing process
DRreadiness
monitoring applicationsPlanning developing and implementing disaster recovery planscan be complex and time-consuming but a new class of apps canhelp you determine if your DR plans are properly synchronizedwith your IT operations BY PAUL KIRVAN
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
DR monitoring applications address one of the key causes of recoveryfailures configuration drift Configuration drift occurs when storage andother IT gear is upgraded or replaced but the DR documentation and pro-cedures arenrsquot updated to accommodate those changes With the DR docsout of synch with the real-world setup recovery efforts are more likely tofail The purpose of monitoring applications is to find those discrepanciesand improve the odds of a successful test or actual recovery
WHAT SHOULD BE MONITOREDFrom a disaster recovery perspective the logical first point of focus wouldbe protecting critical data This means backing up data on-site and off-siteusing a variety of methods including disk-to-disk (D2D) and disk-to-tape(D2T) backup data mirroring to an off-site location or cloud service andshipping backup tapes to an off-site storage facility
The next focal point would be hard-ware and applications You want to ensure these assets are regularlymonitored for proper performance After that but not necessarily in orderof relevance are network-based assets such as local-area networks(LANs) wide-area networks (WANs)storage-area networks (SANs) premises-based systems such as routers andswitches and voice systems such asPBX systems and Voice over IP (VoIP)systems
In a typical data center operationalassets are linked together in a varietyof ways mostly via networking assets The links and relationships amongthe various resources may be complex making it even more difficult to detect configuration drift or other discrepancies
Successful disaster recovery plans are worthless unless theyrsquore regularlyexercised to ensure everything thatrsquos critical to the business is being addressed But the majority of organizations typically exercise their DRplans only once a year or even less frequently
To maintain a careful vigil over your critical data protection activities aswell as systems applications and networks yoursquoll need timely performance
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
34 STORAGE November 2011
Successful disasterrecovery plans areworthless unlesstheyrsquore regularlyexercised to ensureeverything thatrsquoscritical to the business is beingaddressed
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
35 STORAGE November 2011
Issue
General
Existing IT environment
Integration into DR plans
Cloud-based technologies
Network
IT management
Outsourcing and managed services
Questions to askKey points to consider
bull What do you want to monitorbull What information do you want to collect Is it for the entire IT environment
(on-site and off-site) or just partsbull Who will use the monitoring systemsbull Will the system work in a virtualized environmentbull Is output only data or data analyzed against metricsbull Can system find dependencies and other relationships single points of failure
and any other weaknesses before mapping them to your requirements
bull Define a ldquonormalrdquo operating environmentbull Do you want software to look for all possible risksbull Is just flagging unknowns enoughbull How will the monitoring tool integrate with daily operationsbull What data and parameters are essential for optimal knowledge of the backup
environmentbull Does the tool replace or complement existing monitoring appsbull Can existing monitoring systems be adapted modified or upgraded to support
DR monitoring capabilities
bull What results are required to support the DR planbull Define the data yoursquoll need prior to conducting an exercisebull Can the DR monitoring system replace DR exercising How will it supplement
exercisesbull Do overall backup strategies need to be changed with the introduction of
DR monitoring tools
bull If you use cloud storage services how should the DR monitoring system integrate with them
bull Will the system operate outside data center boundaries Will it be used to monitor external applications
bull Will DR monitoring tools enhance the value of cloud services
bull Do DR monitoring systems have to examine network infrastructures or are current systems adequate
bull Will the system have to monitor VoIP systems or older PBX systemsbull Will the monitoring system be used for unified communications call center
environments or audiovideoconferencing systemsbull Will the monitoring tool provide performance and DR data across multiple
communications environments
bull Will the DR monitoring tool be used as part of a normal change management process
bull Are there any limitations related to the number and type of systems that can be analyzed by the DR monitoring tool
bull Conduct a risk analysis of what might happen if no DR monitoring applications were being used
bull Determine the return on investment of a DR monitoring application
bull If you use a third-party service provider (eg hot site firm managed services vendor outsourced data center) does the service offer DR monitoring as a value-added extra or fee-based service
What to ask before buying a DR monitoring application
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
data from all critical assets If you know the health of your IT infrastructurein real-time yoursquoll be better prepared to respond quickly and effectivelywhen something disrupts operations
EXISTING PERFORMANCE MONITORING SYSTEMSIf your organization is in the medium- to large-scale range chances areyou have already invested in a variety of performance monitoring tools You may have applications that scan only one system and others that maymonitor a broad range of activitiessuch as network performance or information security monitors
But if you add disaster recoveryon top of your existing monitoringactivities yoursquoll have to determine ifthe existing performance monitorscan integrate with the DR plans Evenmore important is the need for per-formance data to be optimized for DRplan requirements such as validatingthat data replication activities arebeing completed as planned Andwhat if you have hundreds or thou-sands of applications distributedeverywhere How can you determinethat everything is performing normally If a critical system begins to mal-function it needs to be brought to your attention as quickly as possible
DR MONITORING SYSTEMS EXPLAINEDA relatively new category of software products has emerged that can pro-vide actionable performance data that can be synchronized with DR plansAccording to Jon Toigo CEO at Toigo Partners International there are threedifferent types of disaster recovery monitoring tools ldquoThere are softwareproducts that store information about your plan and create the plan docu-ments Next are tools that set up scenarios to help you fail over from oneset of technology to another set while providing data replication servicesAnd third there are passive tools that monitor the data protection processesrdquoWersquoll focus on the third type of DR monitoring tools
STORAGE
STORAGE November 2011
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
36
Even more importantis the need for per-formance data to beoptimized for DR planrequirements suchas validating thatdata replicationactivities are beingcompleted asplanned
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
To find out if CDI Managed Services are right for you please contact us
888-524-3877A CDI Company
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
As a storage manager your primary concern is likely data protection so yoursquoll want something that monitors all protection-related activities as described earlier You may assume that normal due diligence activitieswould be sufficient and another specialized system wouldnrsquot be necessary
ldquoDespite your due diligence and efforts to build a high-availability datareplica you may still not have an exact duplicate of your production envi-ronmentrdquo said Kathleen Lucey FBCI president of Montague Risk Manage-ment vice president of the Business Continuity Institutersquos (BCI) USA Chapterand vice chairperson of the BCI Global Membership Council ldquoThere couldconceivably be undetected incompatibilities among some components Ifincompatibilities do exist you may not know about them until you switchto the backup site and things donrsquot workrdquo
That same kind of attention mustbe paid to critical IT operational activities such as change manage-ment and configuration managementldquoAre todayrsquos practices in servicemanagement and change manage-ment necessary to ensure the in-tegrity of DR capabilities and plandocumentationrdquo asked Douglas Weldon FBCI an IT executive with a major financial services firm andpresident of the BCIrsquos USA ChapterldquoThe answer is yes these practicesare indeed necessary but additionaltools are needed to fill in the gaps ofon-going monitoringrdquo
ldquoAside from detecting operatingweaknesses the monitoring productshould be able to flag all changes regardless of sizerdquo said Harvey BetanMBCI and president of H Betan Inc aNew York City-based business continuity consultancy ldquoBeing an automatedproduct it can also inspect the IT environment faster than an individualrdquo
In an ideal world data center managers have a single interface thatlinks all monitoring systems and provides a concise integrated dashboardof all infrastructure performance Reports on disaster recovery perform-ance metrics would be one of the outputs
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
38 STORAGE November 2011
ldquoAside from detecting operatingweaknesses themonitoring productshould be able toflag all changesregardless of sizeBeing an automatedproduct it can alsoinspect the IT envi-ronment faster thanan individualrdquo
mdashHARVEY BETAN MBCI president H Betan Inc
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Disaster recovery monitoring systems perform four primary functions1 Data capture and discovery 2 Data compilation 3 Data analysis using predefined configuration data and performance
metrics 4 Data presentation
DR monitoring systems typically connect to their intended systems viainternal (eg LANs) and external networks (eg the Internet) Systemsldquosniffrdquo for specific activities as definedin their logic by sending out speciallydesigned packets to look for specificactivities
Data captured during the discoveryprocess is analyzed according to predefined parameters ldquoThese prod-ucts collect information on applications systems hardware configurationslinks between systems etc to pro-duce a map of the IT infrastructureand the linkagesrdquo Toigo Partnersrsquo Toigosaid ldquoThey can also be integrated withconfiguration management database[CMDB] software for easy referencerdquoThe CMDB stores data about IT infra-structure assets relationships andconfigurations But since it doesnrsquot have analytic capabilities itrsquos difficult toeffectively use that data to protect data and ensure business continuity
SOME EXAMPLES OF DR MONITORING SYSTEMSAptare Inc StorageConsole 8 Fabric Manager StorageConsole 8 is amonitoring system that addresses data storage itrsquos designed to providegreater visibility and management capabilities into SANs Itrsquos part of anAptare portfolio that also includes Backup Manager Capacity Manager Virtualization Manager and Replication Manager The SAN mapping capabilitygives administrators a view of the SAN topology from server to fabric tostorage systems A change management feature performs a dependencyanalysis based on proposed changes to a SAN
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
39 STORAGE November 2011
These products collect informationon applications systems hardwareconfigurations linksbetween systemsetc to produce a map of the IT infrastructure and the linkagesmdashJON TOIGO CEO Toigo Partners International
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
BMC Software Inc Atrium Discovery and Dependency Mapping Thesystem features a library of predefined product configurations (32000-plus updated monthly) reference data for hardware power consumptionand heat dissipation and software end-of-life dates as well as automateddiagnostics that identify the location and cause of discovery issues Soft-ware doesnrsquot have to be installed on discovered devices
Continuity Software RecoverGuard The latest version of this productRecoverGuard 40 leverages data contained in the CMDB by scanning theinfrastructure performing analyses of the information it collects and identi-fying issues that could impact availability recoverability or data protectionRecoverGuardrsquos knowledge base contains more than 2000 gap signaturesand hundreds of potential data protection gaps
Hewlett-Packard (HP) Corsquos Discovery and Dependency Mapping Advanced Edition (DDMA) This system automates discovery and depend-ency mapping of services applications and underlying infrastructure Mapping helps facilitate failure impact analyses which minimize downtimeImproved visibility into the existing IT infrastructure helps reduce opera-tional expense defers capital expense and improves business uptime Ac-cording to HP 80 of all service disruptions are caused by faulty changesThis product can provide visibility for improved change management
VMware Incrsquos vCenter Application Discovery Manager VMwarersquosproduct provides continuous discovery and mapping of applications theirdependencies and configurations The system provides real-time visibilityinto the data center from an application standpoint VMware support enables discovery of application services and configurations in virtual environments and complements VMware vCenter Server by mapping the physical to virtual dependencies
PRE-PURCHASE PLANNING FOR DR MONITORING SYSTEMSAs with any system acquisition successfully implementing a DR monitoringapplication takes a fair amount of planning Perhaps the most importantpart of the process is determining specific needs and what you hope toaccomplish with a product For more tips about evaluating and buying a DR monitoring app see the chart entitled ldquoWhat to ask before buying aDR monitoring applicationrdquo
ldquoDo the basics Make sure the product works and works on your configu-rationrdquo Montague Risk Managementrsquos Lucey recommended ldquoUse change
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
40 STORAGE November 2011
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
control and the monitoring tool to ensure that all production changes arealso concurrently applied to backup configurationsrdquo
Betan from H Betan agreed and added ldquoThe product should also not usetoo many resources that could slow down normal processing Be sure totest the system thoroughly before it enters productionrdquo
NEW TOOLS FOR AN OLD PROBLEMTo ensure your disaster recovery efforts are performing optimally DR mon-itoring tools can help by proactively spotting potential problems beforethey occur ldquoA DR monitoring system can help users achieve several keygoalsmdashthey need to be able to recover and restore their data re-host theapplications and reconnect their usersrdquo Toigo Partnersrsquo Toigo said Added executive Weldon ldquoThese systems are at an early level of maturity andsince each IT environment will be different remember that one size will not fit allrdquo
By identifying potential threats in real-time DR monitoring tools can enhance your ability to respond to and recover from emergencies thusproviding a higher level of disaster recovery and overall preparedness 2
Paul Kirvan is an independent consultantIT auditor and technical writereditoreducator with more than 22 years of experience in business continuity and disasterrecovery
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
41 STORAGE November 2011
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Building Trust Around The Globe
When you want to establish trusted relationships
with anyone anywhere on the internet turn to Thawte
Securing Web sites around the globe with
bull strong SSL encryption
bull expansive browser support
bull multi-lingual customer support
bull recognized trust seal in 18 languages
Offering outstanding value Thawte is for those
who know technology Secure your site today
with a Thawte SSL Certificate
wwwthawtecom
copy 2010 Thawte Inc All rights reserved Thawte the Thawte logo and other trademarks service marks and designs are registered
or unregistered trademarks of Thawte Inc and its subsidiaries and affiliates in the United States and in foreign countries All other
trademarks are property of their respective owners
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
bGet real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
43 STORAGE November 2011
EFORE NOW you might not have considered your IT shop as one that needs ascale-out system But scale-out systemsmdashwhich started out in the network-attached storage (NAS) space because some industries rely on very largefiles that require a lot of bandwidth to meet performance requirementsmdashhave far-reaching implications for a varied number of environments today
For instance several major industries that once operated in traditionalpaper- or microfilm-based modes are finding that their digital data storesare threatening to overwhelm them These are attractive vertical marketsfor scale-out NAS vendors that can provide high-performance applicationsupport
hot spots | terri mcclure
Who really needs scale-out systemsIndustries that once operated in traditional paper-based
models are finding themselves overwhelmed by their digital data stores But scale-out NAS can
provide high-performance application supportSc
ale-u
p
Scale-out
General Purpose Enterprise IT
Consumer and Prosumer
Media andEntertainment
BiotechBioinformatics
Web-Based Applications and
Services
Manufacturingand Design
Oil and Gas
FinancialServices
HPC Workloads(Universities
R amp D)
Price
and m
ulti-u
ser p
erfor
manc
e (fil
e ops
sec
)
Capacity and bandwidth-intensive performance (MBps)Source Enterprise Strategy Group 2011
VERTICAL AFFINITY FOR SCALE-OUT NAS
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
44 STORAGE November 2011
If we look at the throughput vs IO axis model in the ldquoVertical affinity for scale-out NASrdquo graphic these industries have many applications that require the very high throughput the parallel data services found in manyscale-out NAS systems (and coming next year with pNFS) can deliver ex-ceeding the MB per second capabilities of traditional scale-up NAS systems
As recently as five years ago this chart would have looked very differentMany of the workloads in the upper right would have been crowded into theleft-hand side of the chart But advances in processor technologymdashsuch asmulticore processing and much faster chip setsmdashand video graphics anddesign softwaremdashsuch as 3-D CAD 4-D medical imaging and high-definitionTV to name just a fewmdashhave created new types of workloads that demanda very different performance profile They create huge files and multithread-ed requests that a single- or dual-processor scale-up system wouldnrsquot be able to service in a timely manner causing production to slow or the system to time out waiting for the request
Letrsquos take a deeper dive into a few verticals to illustrate my point
Financial services These users who are accustomed to managing extremely large volumes of transactional information are also now heavyusers of high-performance parallelfile systems for efforts such as market-performance forecasting and business intelligence These ef-forts involve files that arenrsquot just bigtheyrsquore also long running compute-intensive and require a high level ofdata protection and immediate dataavailability Financial services users in particular look for scale-out archi-tectures that remove data integrationbottlenecks Data integration is a core task in financial services IT For theseusers an ideal NAS solution is one that performs faster as the number ofnodes increases
Life sciences Not surprisingly organizations engaged in health-relatedscientific discovery are actively interested in parallel file system solutionsoffering high-bandwidth data transfer and massive scalability At these organizations collaboration at an intensive level is typically evident Forexample the IT team may need to find ways to enable sharing of very largegene-sequencing files or proteomic data across thousands of researchers
Financial servicesusers in particularlook for scale-outarchitectures thatremove data inte-gration bottlenecks
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
45 STORAGE November 2011
To be successful these companies must accelerate their discoveryprocesses the faster they develop a new drug the faster it can be testedand approved for use in real-world medical and scientific applications One IT-centric way for such organizations to accelerate a drug discoveryprocess is to use a high-performance parallel file system infrastructurethat never requires disruptive forklift upgrades
Manufacturing and design High-tech manufacturers aerospace com-panies nano-electronics startups CADCAM design firms and many othersalso need tremendous amounts of storage And theyrsquore all looking forways to optimize data management Users in these industries need fault-less capacity expansion to handle digital growth and improve informationsharing among engineering teams Outages are economically damaging inthese environments so users in the manufacturing and design segmentseek to deploy file-based storage that offers near-total reliability and easycapacity upgrades on the fly They look for automation to assist with file-system administration data movement replication and migrationtiering
Media and entertainment The operating model of media and enter-tainment organizations has evolved dramatically In years past they per-haps produced print magazines that are now available in an ldquoonline-onlyrdquoformat Not only does all editorial content need to be quickly available toreaders and content generators butall the advertising files do too Largevideo files are also exacerbating thedata growth problems at digitally intensive media and entertainmentcompanies
Todayrsquos media and entertainmentorganizations are generating and pro-tecting terabytes or petabytes of filedata At some enterprises much ofthe data is created at the ldquoedgerdquomdashinremote news bureaus or CGI designstudios separated from main data centers That operational structure bringsproblems related to data replication for backup and can even impede thedisaster recovery (DR) capability of the infrastructure Media and entertain-ment organizations are looking at high-performance scale-out NAS solutionsto solve a variety of problems for instance to improve the performanceof a virtual server infrastructure or to ensure that information is instantlyand always available to content creators and consumers
Todayrsquos media and entertainmentorganizations aregenerating and protecting terabytesor petabytes of file data
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Oil and gas Uncovering oil and gas reserves was once a guessing gameToday itrsquos a precise scientific endeavor that relies on digitized data Three-dimensional visualization to spot possible resources has become an ever-present tool for the industry as fields decline and extraction operations become more complex IT managers working in the oil and gas vertical market are challenged to find NAS infrastructures that can support thesharing and protection of the huge data sets resulting from oil reservemodelingsimulation work Without an architecture that can maintain performance as data storage capacity grows sustaining a competitiveedge becomes more difficult mainly because the ldquotime-to-resultrdquo (the extraction of the resource) lengthens Scale-out NAS is a good solution foroil and gas organizations dealing with enormous computational simulationsthat in a very direct fashion hold the key to their competitive success
Traditional high-performance computingacademics and researchAstrophysicists molecular biologists chemists nuclear physicists andeven social scientists working in the public sector are heavy generatorsand consumers of data For example at the Large Hadron Collider run byCERN the team in charge of IT was managing 70 PB of storage by mid-2010Even far smaller research facilities (usually working in cost-constraineduniversity settings or commercial labs) rely on high-performance gridcomputing and parallel file system architectures to support modeling andsimulation efforts that could solve real-world problems and answer bigquestions Their work requires low-latency network clusters that can handleextremely intensive performance and bandwidth demands
These industries were the first real adopters of scale-out systems because they absolutely needed the performance capabilities scale-outsystems provide in terms of throughput But a majority of shops shouldrealize the efficiency and operational savings that come with storingmany petabytes of data within a single namespace Thatrsquos why scale-outsystems are finding a home in cloud infrastructures allowing companieslike Gluster (recently acquired by Red Hat)mdashwhich offers a scale-out filesystem that runs on commodity hardware and can support block file andobject datamdashto gain big interest from cloud-based businesses and enter-prises building private clouds Enterprise Strategy Group forecasts that80 of all external NAS systems revenue will come from scale-out systemshipments by 2015 and both ldquobig file datardquo and cloud will be at the core ofthat growth 2
Terri McClure is a senior storage analyst at Enterprise Strategy Group Milford Mass
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
46 STORAGE November 2011
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
DYNAMIC SECURITY FOR THE GLOBAL NETWORK
Do All BAckup AnD RecoveRy SolutionS DeliveR FlexiBility AnD Simplicity conSiDeR youRSelF WARneD
[ Tough quesTion 8]
SonicWALLrsquos all-new Continuous Data Protection (CDP) Series easily and flexibly prevents data loss and downtime with continuous or scheduled point-in-time backup and one-touch restore featuring granular policy enforcement and data de-duplication CDP protects local and remote Windowsreg Mac OSreg or Linuxreg servers desktops or laptops Multiple disaster recovery options include Site-to-Site Backup Offsite Backup Local Archiving and Universal System Recovery With CDPrsquos granular protection administrators can reliably back up only business-related data CDP also lets administrators enforce global policies and schedules or allow users to set their own Data de-duplication technology optimizes storage efficiency by handling more data with less disk space Learn more about the easy-to-use and flexible SonicWALL Continuous Data Protection Series at wwwsonicwallcomwarning
copy 2011 SonicWALL Inc SonicWALL and the SonicWALL logo are registered trademarks of SonicWALL Inc
NETWORK SECURITY
SECURE REMOTE ACCESS
WEB AND E-MAILSECURITY
POLICY AND MANAGEMENT
BACKUP ANDRECOVERY
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
48 STORAGE November 2011
hYBRID CLOUDS HAVE been getting more than their fair share of attention fueledby recent announcements by VMware and other big players Hybrid cloudsbring together public off-premises clouds and on-premises clouds (whichtoday may be nothing more than traditional in-house IT environments) tocreate a more functional solution But the secret sauce of a hybrid cloud isthe connecting technologies that reside in the middle To effectively bridgeoff-premises clouds with on-premises computing and storage environ-ments a hybrid cloud solution should provide enterprise-class securitycross-cloud management workloaddata portability and interoperability
Hybrid clouds offer some compelling advantages and we believe theyshould be the ultimate ldquoend gamerdquo of any organizationrsquos cloud deploymentstrategy Though the market is still developing major hardware and soft-ware vendors are already working to bring hybrid cloud offerings to marketInterestingly the hybrid cloud model is probably further along in the datastorage market where suppliers are already building and delivering hybrid-like cloud storage products
Most current cloud storage use cases are designed to augment existingstorage practices in a companyrsquos data center or private cloud by extendingsome of those practices into a public cloud The first wave of products issquarely focused on overcoming challenges that canrsquot easily be resolvedwithout the help of a cloud The most widely adopted solutions to date arebuilt on gateways that enable backup and recovery archiving and disasterrecovery (DR) Other emerging use cases include cloud-enabled primaryand near-line storage and storage configurations that use the cloud to enhance application scalability availability andor performance
Letrsquos take a look at a VMware backup and recovery use case One of thebiggest pain points for VMware administrators today is backup Given typical
readwrite | jeff byrne
Cloud storage ahead of the packfor hybrid cloud deployments
The idea of turning over storage systems to the cloud hasnrsquot caught on with enterprises but hybrid cloud storage products show how to leverage both in-house and off-site storage
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
VMware consolidation ratios in the range of six to 10 virtual machines (VMs)per host a virtual server administrator must choose a backup approachthat successfully protects each VM without impacting application perform-ance To address this challenge many small- to medium-sized businesses(SMBs) have turned to VMware Data Recovery (VDR) which efficiently backsup VM images to disk by deduplicating data and updating only blocks thathave changed since the last backup
Using VDR to streamline VMware backup is a great first step but whathappens in the event of a prolonged outage or disaster Most SMBs canrsquotafford to set up and maintain a remote physical DR site but most coulduse the cloud
TwinStrata Incrsquos CloudArray is a block-based cloud storage gate-way that leverages native iSCSIand is packaged as either a physi-cal or virtual appliance for VMwarevSphere Microsoft Hyper-V or CitrixXenServer environments VMwareData Recovery sends deduplicatedbackup data to a CloudArray appli-ance which is stored in thin provisioned virtual volumes Data is written tothe CloudArray cache (with the level of caching configurable on a per-volumebasis) and then replicated to cloud storage At-rest and in-flight encryptionensures that data is secure CloudArray supports all major public clouds orprivate clouds
In the event of a prolonged outage frequently accessed data is immedi-ately restored from cache and data in the cloud is replicated back to theCloudArray appliance either at a VM or file level Data recovery from thecloud can also be automated and orchestrated by employing VMwarevCenter Site Recovery Manager This approach provides rapid data recoverywithout the burden of supporting a remote site If you currently use tapefor data archiving this approach can result in even bigger savings sinceyoursquoll no longer have tape maintenance and logistical costs
Riverbed Technology offers a similar type of cloud storage gateway solu-tion via its Whitewater appliance but itrsquos file based rather than iSCSI basedWhitewater presents itself to popular backup applications as a CIFS or NFSdisk With disk space of 2 TB to 8 TB Whitewater can cache recent backupdata locally for rapid recovery The appliance takes advantage of RiverbedrsquosWAN acceleration technology and built-in deduplication to replicate data
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
49 STORAGE November 2011
Most SMBs canrsquot affordto set up and maintaina remote physical DRsite but most coulduse the cloud
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
to a public storage cloud such as Amazon S3 ATampT Synaptic Storage as aService (based on EMC Atmos) or Nirvanix Cloud Storage Network As withthe TwinStrata product no hardware is needed in the cloud and data is fullyencrypted both in motion and at rest
With TwinStrata CloudArray and Riverbed Whitewater the cloud becomesanother storage tier providing backup and recovery archiving or full DR pro-tection These innovative products take advantage of cloud storagersquos greateststrengths pay-as-you-go economics coupled with scalability and agility
Letrsquos look at a completely different cloud storage use case focused onincreasing Microsoft SharePoint scalability availability and performanceSharePoint is widely used in busi-nesses of all sizes but it has somearchitectural limitations that makeit less effective as it scales to cover more data and more usersIn particular the inefficient stor-age of objects in Microsoft SQLServer database tables can lead toperformance and data protection issues as well as difficulties in implementing high-availability (HA) and DR solutions
StorSimple Incrsquos SharePointDatabase Optimizer product wasdeveloped to address these issues The product consists of a StorSimpleappliance and Microsoft framework plug-in for SharePoint servers Ratherthan storing file objects (BLOBs) in SQL Server which reduces efficiencyand drives up capacity requirements and cost the SharePoint Database Optimizer first deduplicates and compresses the file objects and then usesEBS or RBS BLOB interfaces to store the objects in encrypted form in a publicor private cloud Database sizes are reduced by up to 95 allowing the BLOBvolumes to take advantage of cloud economics
On the other hand the StorSimple appliance initially stores SharePointmetadata database volumes on built-in solid-state drives (SSDs) and usesa Weighted Storage Layout (WSL) mechanism to automatically tier contentbased on usage patterns moving infrequently accessed metadata volumesto the cloud
StorSimplersquos SharePoint approach has several major benefits including
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
50 STORAGE November 2011
With TwinStrataCloudArray andRiverbed Whitewaterthe cloud becomesanother storage tierproviding backup andrecovery archiving or full DR protection
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
dramatically greater scalability and performance far more efficient use ofstorage capacity and lower overall costs And by separating metadata fromfile objects and streamlining their storage StorSimple accelerates backupsand enables cost-effective HA and DR protection
We believe these three sample products are great examples of the out-of-the-box thinking that characterizes a growing number of emerging hybrid-like cloud storage offerings And we anticipate a steady stream of cloudstorage innovations coming in the year ahead 2
Jeff Byrne is a senior analyst and consultant at Taneja Group He can be reachedat jeffbyrnetanejagroupcom
STORAGE
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
51 STORAGE November 2011
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Tape still plays a role in the data centerLETrsquoS SKIP the ldquotape is deadrdquo rhetoric and get down to facts 58 of Storage readersuse tape as much or more than they did three years ago and only 16 have ban-ished tape Thatrsquos not to suggest disk isnrsquot bigger than ever in backup but itrsquos noteven close to sending tape to the showers for good Forty-nine percent of respon-dents use disk more in their backups than they did three years ago while 11 donrsquotuse any disk at all in their backup ops Regardless of the mix of disk and tape 80say that eventually all or some of their backup data winds up on tape Most firms(60) will leave backups on disk for 30 days or less before spinning them off to tapeTape clings to its backup role but itrsquos not the only one itrsquos good at 37 use tape fornon-backup apps like archival (83) and nearline storage to augment disk (28) Itdoesnrsquot look like the tables will turn too much for tape next year as one-third ofthose surveyed expect to buy some tape gear in the coming year mdashRich Castagna
ldquo The economics of tape vs disk still preclude the exclusive use of one technologymdashrather disk provides a beneficial front-end cache to long-term tape storagerdquo
mdashSurvey respondent
snapshot
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
52 STORAGE November 2011
80
Compared to three years ago how would you characterize your companyrsquos use of tape
Say all or some of their data backed up to disk
eventually gets copied to tape
11
8
21
60
Longer than 90 days
60 days to 90 days
30 days to 60 days
Less than 30 days
0 10 20 30 40 50 60
How long does your backup data stayon disk before being spun off to tape
What tape equipment will your company purchase
within the next 12 months
68 We will purchase new tape drives
40 We will purchase at least one new tape library
40 We will purchase new robotics or other tape library equipment
Respondents could make multiple choices
We donrsquot use any tape in our backup process
We still use tape but less than we did three years ago
Our use of tape hasnrsquot changed much in three years
We use tape more now than we did three years ago
26
3226
16
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds
bull Fluid Data Storage Driving Flexibility in the Data Center - Eight Must Have Technologiesfor the IT Director
bull 7 Ways Compellent Optimizes VMware Server Virtualization
SPONSOR RESOURCES
See ad page 7
bull Improving Backup and Recovery for VMware vSphere Environments
bull Backup and Recovery of Large Scale VMware Environments
See ad page 32
bull Confidently maximize virtual investments with IBM Integrated Service Management
bull Analyst Whitepaper Storage-efficient Data Protection and Retention
See ad page 26
bull Cloud Tape Hybrid Data Protection Best Practices
bull Guide to Improving Your Tape Storage Practices
SPONSOR RESOURCES
See ad page 10
bull Deduplication for Dummies
bull Learn how data deduplication technology works and how Quantumrsquos disk-based backupsolutions can help your data center exceed expectations
See ad page 20
bull High performance scalable Network Attached Storage solution for the enterprise
See ad page 42
bull Extended Validation SSL Certificates
bull Securing your Online Data Transfer with SSL
SPONSOR RESOURCES
See ad page 47
bull SonicWALL Continuous Data Protection Video Guide
bull Five Best Practices to Optimize Backup Management for Effective Data Recovery
bull SonicWALL CDP Case Study Video Continental Group
Cover
November Table of Contents
Lets get real about the cloud
Time is right for SSDs
Storage pros pay still on the rise
NAS system buying decisions
DR readiness monitoring applications
Who really needs scale-out systems
Cloud storage ahead of the pack for hybrid cloud deployments
Tape still plays a role in the data center
Editorial mastheadDecember 2011 preview
Sponsor resource page
Get real aboutthe cloud
Time is right for SSDs
Salary survey
NAS systems
DR readinessmonitoring
Scale-out systems
Hybrid clouddeployments
Tale of the tape
Sponsorresources
53 STORAGE November 2011
STORAGE
COMING IN
DecemberHot Storage Technologies 2012Storage mag and SearchStoragecomeditors put their heads together andpredict the top new techs that willhave the most profound impact onstorage environments over the next
12 months We also grade our choicesfrom the previous year See what storage techs should be on your
must-have list for 2012
Mobile Backup Therersquos an App for That
Users might have a significant amountof their firmrsquos intellectual property onultra-portable devices Some vendorsoffer apps to protect that data and
ensure it stays within corporate confines We look at whatrsquos availablehow well it will work and whatrsquos still
needed to protect mobile data
Quality Awards VI Tape LibrariesWho said tape is dead Certainly notthe legions of users who value tapersquoseconomy portability and usefulnessfor long-term storage of backup andarchive data Our Quality Awards pro-gram surveys users of midrange and
enterprise tape libraries about qualitytechnical support features and otherkey servicereliability considerations
And donrsquot miss our monthly columnsand commentary or the results of
our Snapshot reader survey
TechTarget Storage Media Group
Vice President of Editorial Mark SchlackEditorial Director Rich Castagna
Senior Managing Editor Kim HefnerExecutive Editor Ellen OrsquoBrien
Creative Director Maureen JoyceContributing Editors
Tony Asaro James Damoulakis Steve Duplessie Jacob Gsoedl W Curtis Preston
Executive Editor Ellen OrsquoBrienSenior News Director Dave RaffoSenior News Writer Sonia R Lelii
Senior Writer Carol SliwaSenior Managing Editor Kim Hefner
Assistant Site Editor Rachel KossmanEditorial Assistant Hillary OrsquoRourke
Executive Editor Ellen OrsquoBrienAssistant Site Editor Rachel Kossman
Senior Site Editor Andrew BurtonManaging Editor Ed Hannan
Assistant Site Editor John HilliardFeatures Writer Todd Erickson
Senior Site Editor Sue TroyAssistant Site Editor Francesca Sales
Senior Site Editor Sue TroyUK Bureau Chief Antony Adshead
Assistant Site Editor Francesca Sales
TechTarget Conferences
Director of Editorial Events Lindsay JeanlozEditorial Events Associate Jacquelyn Hinds