IBM System Storage™ N series – Unified Storage Solutions © 2007 IBM Corporation An Introduction to IBM System Storage N series Name title October 23,

Slide 1© 2007 IBM Corporation
An Introduction to
Name
title
*
Goal of this presentation:
Introduce potential customers to IBM N series in such a way that N series is in the customer’s consideration set for every data management purchase.
Script suggestions can be found in the notes section of each slide. To print these pages, use File/Print to access the “Print” dialog then, from the “Print What” drop down in the lower left section of the dialog box, select “Notes Pages”.
Please see the Corporate Pitch Presentation Guide for more information, supporting data, and suggestions for additions and variations to the presentation.
Last updated: Jan 3, 2007
Changes from previous version:
* new Gartner Magic Quadrant and Speaker’s notes
* updated references to include FY07 Q2 numbers
IBM System Storage™ N series – Unified Storage Solutions
© 2007 IBM Corporation
October 23, 2007
Summary
*
October 23, 2007
In April, 2005, IBM and Network Appliance announced a strategic storage relationship to drive information on demand solutions and to expand IBM's portfolio of storage solutions.
IBM agreed to OEM (rebrand with IBM logo, service, support and warranty) the entire storage portfolio
+
=
1.unknown
October 23, 2007
Explosive data growth
Data security and compliance
*
Goal of this slide: Open dialog with customers to explore the challenges they are most concerned with, allowing you to better position our solutions with the customers requirements.
Script:
We talk with lots of customers, and even though their environments and businesses are very different, many of the same challenges come up over and over again. IT managers consistently struggle with handling explosive data growth, the need to do more with less, keeping up with change while maintaining constant availability, keeping data out of the hands of the wrong people and keeping their bosses out of jail.
One of the newer trends we’re seeing is a real focus on application integration, because people want to spend more of their resources focused on increasing the performance of their critical business applications, not worrying about low-level infrastructure components like storage. This is one of the reasons that we work so closely with key vendors like Oracle, Microsoft, and SAP, so that their applications can take direct advantage of our innovation. Over time, that integration will deliver the benefits of our storage innovation directly into the hands of their customers.
[Click mouse to build]
Another trend we’re seeing is the recognition of the broad impact a company’s IT infrastructure can have on business. This, in turn, means CIOs everywhere are looking to align their IT capabilities with the company’s business challenges.
These are just some of the things we see broadly across the industry, but I’m curious about the challenges you face. Do they align to this list pretty well? What would you say are the biggest and hardest challenges you face?
(Use this time to open up dialog.)
*
October 23, 2007
Manage
Protect
Retain
*
Goal of this slide: Associate customer challenges with the N series product portfolio.
Script:
At the highest level, we focus on addressing 5 business challenges:
Storing Data
Managing Data
Protecting Data
Retaining Data, and
Ensuring Business Success
In the Storage area, we deliver the broadest scalable storage platform in the industry, built on a Unified Architecture that truly delivers unmatched administrative simplicity.
Running across that entire storage platform is a suite of high-productivity data management tools, designed to increase the productivity and availability of your business applications.
We also have an industry leading portfolio of disk-based data protection solutions that not only speeds your backup and recovery but does so at a cost that allows you to protect more of your data.
The combination of amazing data growth and increasing regulations increases the need for flexible and affordable data retention solutions. IBM N series delivers. (add)
*
October 23, 2007
*
Goal of this slide: Demonstrate the difference and consequential value of the N series Unified Architecture approach.
Script:
One of the best examples of N series innovation is the Unified Architectural Model which provides the foundation for the dramatic differences in value N series is able to provide.
First let’s look at the hardware platform model. No matter which of our competitors you look at, they all use the same approach—specialized, incompatible platforms for different functions. They may have a platform for low end, another for mid-range, yet another for high-end, and still another for compliance. Each of these platforms, while robust in its own area, forms an information silo, and an investment dead-end. By contrast, N series systems base solutions on one, extremely broad, scalable and fully compatible platform, totally eliminating the notion of information silos. And to help you get the most out of your investment dollars, every system can be easily upgraded without migrating the data.
This model starts to get even more compelling when you look at the software and processes required. The specialized hardware platforms each run their own, incompatible software, each with its own set of processes and “best practices”. In contrast, the N series family all runs the same set of software, with the same processes. So much so that we hear customers say that they only have to test an application with one N series system—they know what works on one, will work on all.
Add to this the people side of the equation, and you see that all those incompatible platforms each need their own experts, and getting them to work together requires even more people and expensive integration services. With n series, your people need less training, spend less time on making things work together, and because they’re familiar with the systems, they make fewer mistakes—the leading cause of downtime.
I hope this helps you understand how simple concept like “architectural simplicity” can make a big impact to your bottom line.
Delivering Value Across the Datacenter
*
August 28, 2007
N series Gateways
Leverage existing Storage Assets while introducing advanced N series Software functionality
N3700
16TB
N5600
504TB
N5200
84TB
N7700
840TB
N7900
1,176TB
N5500
168TB
N5300
336TB
N3300
68TB
N3600
104TB
Storage solutions that deliver simplified data management for your entire Datacenter
An Introduction to IBM System Storage N series
*
From a hardware perspective:
Let me tell you the base line or (it’s a given that..)
Reliable—Designed to address the needs of business- and mission-critical applications through high data availability and system-level redundancy features
Application Availability – IBM N series systems with Data ONTAP® enable application-level recovery in minutes, not hours, upon failure or user error.
Versatile—Single, integrated architecture designed to support concurrent block I/O and file serving over Ethernet and Fibre Channel SAN infrastructures
Performance - Delivers high, consistent performance for the largest applications and storage consolidation.
Fast—Supports high throughput and fast response times for database, e-mail, and technical applications
Flexible—Fibre Channel and SATA disk-drive capabilities allow for deployment in multiple solution environments, including data retention, near-line storage, disk-to-disk backup scenarios and high-performance, mission-critical I/O intensive operations
Scalable – Non-disruptive capacity expansion and software enhancements
Simple – Easily installed, configured, managed and maintained.
Total Cost of Ownership - Lower acquisition and operational cost than traditional large-scale storage
So which system – how big and how fast do you want it to be?
Seamlessly manage your data from cradle to grave
*
October 23, 2007
*
The N series Advantage
N series storage solutions seamlessly manage data from cradle-to-grave with one OS and a single HW platform to provide
Strong data migration solutions for seamless integration into current data environments
Flexibility and scalability for future growth
Robust Consolidation, Data Protection, and Archival solutions for UNIX and Windows environments
Tightly integrated and Application-aware software for policy-based automation of data protection and disaster recover for VMware, Exchange, Oracle, SAP, SharePoint server & more
*
October 23, 2007
Current Hardware Portfolio
October 23, 2007
*
All N series Systems support the same unified storage features:
High availability features
Hot Swap, redundant components; power supplies, fans, disk drives and storage controller clustering
Connects to NAS, IP SAN and FC SAN environments
Concurrently
FC drive options
SATA options can be used for
Disk backup options (near-line storage)
Archive storage
NAS & IP SAN
*
Single Operating system
System management tools
Storage management tools
High availability features … hot swap and redundancy
Disk Drives: SAS, FC & SATA, multiple disk sizes & speeds … intermixed on all systems**
Non-erasable, Non-rewritable data protection
45+ advanced software features and functions *
Gateways – extend IP community and N series functionality by creating a virtualized N series environment
*
250GB, 7.2K rpm
500GB, 7.2K rpm
750GB, 7.2K rpm
2Gbps Fibre Channel Disks
*
October 23, 2007
*
Cluster controller are designed to offer redundancy, higher availability and increased performance
Compact 3U drawer scales to 4TB*
Three addition expansion units allow a maximum of 16TB*
Single platform offers Unified Protocol support
NAS, iSCSI (IP SAN) & FCP (FC SAN) protocol attachment to Windows, Unix and Linux environments
Enabler for IP-based SANs
Rock solid foundation offers redundant hot swappable components and robust copy services
Suitable for lights-out operation in distributed environments
Extremely suitable for SMB mission critical environments
N3700
N3700 Hardware Overview
N3700 available in single or dual storage controller appliance configurations
Cluster controller configurations contain two controllers inserted into rear of N3700
Cluster configuration are designed to offer the benefits of increased performance with active-active configuration.
Single controller can be upgraded to cluster
Based On Broadcom® MIPS-based architecture
All available adapters ship with base unit
Two 10/100/1000 copper Ethernet
Two Fibre Channel adapters
Protocol Support
Microsoft® CIFS
October 23, 2007
NVRAM Battery
October 23, 2007
October 23, 2007
*
October 23, 2007
Unified Storage: iSCSI, NAS, Fibre Channel
Each Controller: Dual Gigabit Ethernet Ports and Dual 4Gbps Fibre Channel Ports
Onboard Remote Platform Management
2U High, 12 Internal SAS or SATA Drive Bays No PCIe Slots Up to Two External Disk Expansion Units 4Gbps Fibre Channel or 3Gbps SATA Up to 40 disk drives (system total)
4U high, 20 Internal SAS or SATA Drive Bays 1x PCIe Slot per Controller Up to Six External Disk Expansion Units 4Gbps Fibre Channel or 3Gbps SATA Up to 104 disk drives (system total)
IBM System Storage N series Overview
*
This slide is to illustrate the key specifications of the N3300 and N3600 systems.
Key points:
12 (N3300) or 20 (N3600) SAS or SATA disk drives are supported in the controller chassis.
Like all N series systems the N3300 and N3600 provide multiprotocol support
Gigabit Ethernet and 4Gbps Fibre Channel ports are standard on both systems.
On board remote platform management is standard on both systems, like the N5300, N5600 and N7000 series systems.
Unified Storage Model
One architecture for SAN and NAS storage
Unique ability to consolidate file and block storage on single system
Can be used for either primary or secondary storage
*
October 23, 2007
*
The N3300 and N3600 can handle both file and block data in the same system. For block data we have FCP and iSCSI protocols. And for file services we support NFS and CIFS protocols. Gigabit Ethernet an 4Gbps FC interfaces are built into the N3300 and N3600 systems.
N3600 fills gap between N3700 and N5200
Notes: (1) All specifications are for an active-active configuration; (2) max ports combines integrated ports with I/O cards;
(3) maximum N5200 Fibre Channel ports requires 4-port 4Gbps FC adapter card
IBM System Storage N3000 series Sales Training
February 12, 2008
*
Comments:
This slide provides a deeper dive into the specifications of the N3300 and N3600.
Key Points:
Both the N3300 and N3600 use the latest disk technologies and provide both Gigabit Ethernet and 4Gbps FC ports in the base configuration.
Prior to the N3600, there was a significant gap between the N3700 and N5200 in capacity, performance. As well as technological differences like the N3700’s 1Gbps FC back-end storage support.
October 23, 2007
*
October 23, 2007
IBM N5000 series
Unified Storage System
Outstanding price-performance value
Superior application availability
*
Additional Comment:
The N5300 and N5300 Gateway use the same hardware and the same version of the Data ONTAP operating system.
The N5600 is the IBM branded version of the NetApp FAS3070. The N5600 offers customers some very compelling benefits.
One, its unified storage architecture provides multi-protocol heterogeneous platform and is an excellent choice for customers with midrange unified storage needs. Especially those clients who have both IP, iSCSI and FC storage needs and are looking to optimize their IT or consolidate to fewer storage devices.
Two, as you will see in later slides, the system has outstanding value in price/performance.
Three, With all the software capabilities available on the N5600 and all N series systems, it is a very versatile system. This versatility is one of the greatest strengths. With the amount of software options available on N series, you can easily tailor each system to your client’s unique needs.
October 23, 2007
Capabilities N5000 series
Scales up to 504 drives & 504TB capacity
Storage tiers with FC and SATA drives
High-bandwidth I/O design
Integrated remote management
N5300 and N5600
*
Comments
Note that remote management and 4Gb FC have become a standard features in the 2nd generation N5300 and N5600 systems.
10Gb Ethernet is available as a single-port PCI-X card for N5200 and N5500 systems today. The N5300 and N5600 do not have PCI-X expansion slots. Instead, they have higher bandwidth PCI Express (PCIe) slots.
By supporting both FC and SATA disk drives it an ideal system for both primary and secondary storage client requirements.
I would like to point out the N5600 is also based on the 64 bit architecture, and that high bandwidth architecture which is the same as in the N5600 and N7000 series.
The N5600 has onboard Gigabit Ethernet and 4Gbps Fiber Channel ports. For expansion, the N5600 expansion slots are PCI express or PCIe.
October 23, 2007
Versatility: Applications & Environments
Environments
Technology development facilities
*
Comments
IBM N series storage systems are incredibly versatile and the N5000 series is no exception in that regard.
October 23, 2007
*
All specifications shown here are for active-active, dual controller HA configurations. See the detailed technical specifications document for single-controller specifications.
Each of these systems has 8 integrated Fibre Channel ports and 8 integrated Ethernet ports in an active-active, dual controller configuration. The integrated Ethernet ports all run at Gigabit speed. The integrated Fibre Channel ports in the N5200 and N5500 run at 2Gb/sec while the N5300 and N5600 have 4Gb/sec speed on their integrated ports.
The maximum port figures for FC and Ethernet are achieved via a combination of integrated ports and quad-port FC adapter cards or quad-port GbE NICs. For example, a N5300 configured to have 32 FC ports (integrated + adapters) will have 8 integrated Ethernet ports. Conversely, the same N5300 can be configure to have 32 Ethernet ports (integrated + NICs) and 8 integrated 4Gb FC ports.
October 23, 2007
*
*
N7900
N7700
*
Today, we will be talking about a new platform in N series Unifies Storage Family: The 2nd Generation N7000 series. There are 4 key points that we will stress in this presentation :
The new N7000 series scales to massive capacity rivaling frame array storage systems
Data ONTAP and the N7000 series enable storage consolidation that is highly efficient, in terms of both cost and flexibility
N series not only meets the data center baseline for availability features BUT also offers some capabilities that are unique in the storage business
SAN has always been a great strength for IBM;
In fact, the N series family offers the largest portfolio of comprehensive solutions in all of IBM storage lines
And the New N7000 series area 4Gbps end-to-end Fibre Channel
16 x FC and 12 x GbE onboard*
Occupies 12 rack units*
Up to 48 Fibre Channel ports**
Up to 52 Gigabit Ethernet ports**
Built-in enterprise-class manageability
February 12, 2008
* Enterprise active-active configuration.
IBM System Storage N7000 series Sales Training
*
*
February 12, 2008
* Data ONTAP 7.2.4 required to reach capacity points on N7600 and N7800.
N7600 (A/A)
N7700 (A/A)
N7800 (A/A)
N7900 (A/A)
Processing Cores
Max Spindles
*
Similarities Between Systems
October 23, 2007
*
The major differences between the N3000, N5000 and N7000 models
Performance
Scalability
Price
Software is not based on TB’s but based on actual models number
October 23, 2007
*
Data ONTAP™ Features and Benefits
Operating System highly optimized and finely tuned for a single purpose – Serving Storage
Multi-protocol system designed to provide the maximum in investment protection and risk reduction
Usability aids and appliance like features could possible be an IT manager’s dream
Easy to install & quick deployment accelerates time to market
Reduced cost, advanced protection without performance compromises → business continuance
Allows consolidation of storage and servers providing maximum ROI and reduced TCO
Helps improve end-user productivity through increased file sharing solutions
Designed to greatly improved asset utilization with dynamic thin provision
Quick & easy data recovery with proven foundation using technologies.
October 23, 2007
*
Windows, Unix and Linux
Windows user views files over Common Internet File System (CIFS)
Unix user views files over
Network File
*
CATIA (Dassault Systemes) is a classic illustration of the benefits a good multiprotocol implementation can bring to an application.
Multi-protocol: Capacity-savings example
Designed to reduce costs by de-duplicating storage and simplifying system administration
Designed to increase productivity by sharing a single file seamlessly and securely
Designed to eliminates data migration costs by permitting today’s iSCSI LUN to be tomorrow’s FCP LUN
Eases deployment of multiprotocol applications, e.g., Dassault Systemès CATIA
October 23, 2007
Windows Data
UNIX Data
20 TeraBytes
20 TeraBytes
Traditional Approach
*
*
the “non-correctable error rate”
10% - ATA
Assumes 8-drive (7+1) RAID group
Error Rates
Industry-wide trend
As disk drive sizes get larger, the bit error rates increase (per drive)
Non-correctable error rate is uncomfortably close to bits read during reconstruction
October 23, 2007
Bits per RAID Group
*
Designed especially for SATA drive technology
Designed to survive any 2-disk-failure scenario
Breakthrough innovation:
Better protection (>4,000 MTTDL)
Same capacity overhead (typically 1 parity per 6 data drives
Designed to outperform other double-parity” offerings
Routinely on during benchmarks
Combined with SyncMirror (RAID1), the N series products are designed to survive failure of any five disks in one disk protection group
P
DP
RAID-DP
October 23, 2007
P
*
http://www.usenix.org/publications/library/proceedings/fast04/tech/corbett/corbett.pdf
October 23, 2007
0%
5%
10%
15%
20%
*Media/Bit Error + 2nd Failure Likelihood – Double Parity
(during reconstruction of a 16-drive RAID-DP set)
Protected with
(per year)
(full capacity transfer 300GB FC / 320GB SATA)
*Media/bit Error Likelihood – Single Parity
(during reconstruction of an 8-drive RAID 4/5 set)
Industry Statistics: Drive Replacements
*Source: IBM / Seagate / Maxtor / Hitachi & Network Appliance
“… Error Rate per Bits Read”
*
NetApp RAID-DP offers the highest level protection with the best performance available to protect from data loss double resulting from media failure within the same RAID group
Now consider a storage array. Drives are grouped in RAID sets. RAID helps to build resiliency to individual disk drive failures. Upon a drive failure, the RAID set can reconstruct the lost drive using mathematical redundancy built into RAID. The reconstruction requires that all of the bits in the RAID disk drives are read. Data loss occurs when you encounter a bit error during reconstruction read operations.
You now have the three ingredients for a perfect storm under single parity RAID:
Increased (up to 2X) drive failures = more reconstructions with ATA drives
Lower bit error resiliency on ATA drives = increased likelihood of bit errors
Larger ATA disk drives = larger number of bits in a RAID group = increased likelihood of bit errors.
NetApp effectively eliminates this risk with RAID-DP. Others can too (with RAID6). The difference is that our solution creates minimal performance impacts and is extremely simple to deploy.
*
FTP
Snapshot
SnapMover
SyncMirror
FilerView
FlexVol
FlexShare
SecureAdmin
AutoSupport
NearStore feature
Protection Manager
*
Network Access Protocols
CIFS – Windows client & server attached to storage via IP network utilizing file system protocols
NFS – UNIX/Linux client & server attached to storage via IP network utilizing file system protocols
iSCSI – Windows/UNIX/Linux server attached to storage via IP network utilizing block I/O protocols
FCP – Windows/UNIX/Linux server attached to storage via fiber channel network utilizing block protocols
Data protection & Business Continuity
On board copy services via Snapshot & SnapRestore – virtual file & volume copy
Outboard copy services via SnapVault, SnapMirror, SyncMirror
Double parity RAID provides enhanced data protection for SATA drives
Cluster Failover – between redundant nodes
SnapLock & LockVault– non-erasable non-rewritable data protection
System tools, Usability Aids, Provisioning
FilerView – overall system monitoring and management
SnapManager – Exchange & SQL environments
FlexClone – database cloning
FlexVol – Thin Provisioning
Storage consolidation
Server consolidation
Catia migrations
Unified storage
Corporate Compliancy
October 23, 2007
Seamlessly manage your data from cradle to grave
*
October 23, 2007
*
N series Gateways
October 23, 2007
*
Target
Side
NAS
(File)
SAN
(Block)
Host
Side
Departmental
iSCSI
Enterprise
Enterprise
SAN
NAS
Departmental
Corporate
LAN
IBM
*
Unix
Windows
Linux
Appliance (A10 & A20)
Gateway (G10 & G20)
Storage
Disk Storage is provided by external RAID subsystem
Operating System
Data ONTAP
Data ONTAP slightly modified to remove RAID. RAID is dependent on attached storage subsystem
Software
Equal to Appliance models with exception of: Disk Sanitization, SnapLock, Compliance, LockVault Compliance, A-SIS, RAID-DP, RAID4
Performance
FC
FC
*
Tiered Heterogeneous Storage
Line Up for N series Gateways
N5500
Gateway
N5200
Gateway
N7600
Gateway
N7800
Gateway
N series Gateways: Aggregates heterogeneous SAN storage
for file or block access with the compelling data management capabilities of Data ONTAP® and robust software suite
N5300
Gateway
N5600
Gateway
N series R7 Announcement Overview - Analyst Briefing
*
Comments
The IBM N series Gateways enables customers to preserve their investment in existing storage while getting the benefits of Data ONTAP and the N series Manageability Software Family.
N series Gateways offer a low cost way of leveraging FC SAN resources through inexpensive Ethernet connectivity using iSCSI protocols. In addition, SAN resources enabled with N series Gateways can offer enhanced capabilities by leveraging existing resources and SAN capacity for Unified Storage needs (Block and File I/O either through FCP, iSCSI and/or File protocols like CIFS, NFS, HTTP, FTP, etc.
*
Transform Legacy Storage
N series unifies SAN and NAS as well as virtualizes SAN storage from multiple vendors into a single pool for file or block access to provide the benefits
of Data ONTAP and the N series manageability software family.
February 12, 2008
Tiered Heterogeneous Storage
*
N series can help you get more from your current storage investments. The N series Gateways transforms legacy storage systems from IBM, Hitachi Data Systems, HP, EMC, Fujitsu, and 3PAR to provide the benefits of Data ONTAP and the N series manageability software family. The unified and virtualized storage is designed to be easier to provision and manage. Backup time, space requirements, and cost can be dramatically lower with N series Snapshot™ copies. You can also increase storage utilization and reduce costs with N series’ FlexVol thin provisioning. FlexVol can also increase performance by spreading data across more disks. And, with N series controllers in front, your legacy storage can be used not only for FC-SAN but also for NAS and IP-SAN (iSCSI).
An N series Gateway is a disc-less controller with Data ONTAP.
The N series Gateways currently supports the disc arrays of the vendors whose logos you see below.
Here’s a more in-depth support chart for you to inspect:
N series Gateway’s Unified Storage Architecture Enhance SAN Capabilities and Enables IP users Access to SAN Resources
Multi-Vendor SAN Storage
File and/or block access
Aggregation of unused capacity
Advanced data management of
N series R7 Announcement Overview - Analyst Briefing
*
Clients & Servers
Clients & Servers
Clients & Servers
FC SAN
FC SAN
*
Unix
Windows
Linux
A single Gateway supports Unix, Linux and Windows file protocols
Potentially reducing the number of servers required for accessing SAN storage in mixed protocol environments
Connecting small inexpensive servers to premium priced SAN storage is very costly.
FC HBAs, FC switch ports… costs could exceed the actual cost of server
A single Gateway extends the reach of SAN storage to many servers attached to an IP network
Potentially reducing overall connection costs
Allowing access to SAN storage at extended distances
A Gateway provides functionality of N series technology to other vendor storage subsystems
A Gateway provides increased protection investment in SAN storage by increasing SAN utilization & accessibility
N series Gateway
FC
*
A Comprehensive Storage Solution
Proven data management applications for multi-vendor, tiered storage
Customer Benefits
Lowers storage management and operating costs
Increases storage utilization
Enables improved management of conventional storage systems
*
October 23, 2007
*
N series Software
*
October 23, 2007
CIFS Protocol
Allows Microsoft Windows servers & clients access over the IP network using CIFS file system protocols.
Microsoft Windows client access licenses (CAL) are not required. Supports an active directory environment
Allows UNIX and Linux servers and clients access over an IP network using NFS file system protocols.
V2, V3, V4 supported
October 23, 2007
iSCSI Protocol
Allows transfer of data between storage and servers in block I/O formats (iSCSI protocol).
Enables the creation of IP SANs for optimizing the transfer of database traffic in IP environments.
Allows transfer of data between storage and servers in block I/O formats utilizing FCP protocols
Enables participation of an N series storage solution within Fibre Channel SAN environments
Fibre Channel Protocol
October 23, 2007
FTP Protocol
File Transfer Protocol (FTP) is a conversational protocol used to transfer files from an N series running Data ONTAP to a client/user over TCP/IP
It can transfer simple text files and convert between different native formats, such as from a Windows text format to a UNIX text format, or it can transfer raw binary data, such as programs and graphical images.
HTTP (Hypertext Transfer Protocol) is the foundation protocol of the World Wide Web. It sets the rules for exchanges between browser and server. It provides for the transfer of hypertext and hypermedia, for recognition of file types, and other functions.
N series supports HTTP and transfer files from an N series running Data ONTAP to a client/user over TCP/IP
HTTP Protocol
October 23, 2007
FilerView®
A web-based administration tool that allows IT administrators to fully manage N series systems from remote locations.
Simple and intuitive web-based single-appliance administration
SecureAdmin is standard feature of Data ONTAP that enables authenticated, command-based administrative sessions between an administrative user and Data ONTAP over an intranet or the Internet.
SecureAdmin can be used to authenticate both the administrative user and the N series system, creating a secure, direct communication link to the N series system.
It helps protects administrative logins, passwords, and session commands from "cleartext" snooping by replacing rsh and telnet with the strongly encrypted SSH protocol
SecureAdmin™
October 23, 2007
Disk Sanitization
Disk sanitization is the process of physically obliterating data by overwriting disks with specified byte patterns or random data in a manner that helps prevents recovery of current data by any known recovery methods.
This feature enables you to carry out disk sanitization by using three successive byte overwrite patterns per cycle and a default six cycles per operation
FlexShare gives administrators the ability to increase processing utilization without sacrificing the performance of critical business needs.
Allows administrators to consolidate different applications and data sets on a single storage system.
FlexShare gives administrators the control to prioritize applications based on how critical they are to the business and provides a priority mechanism to give preferential treatment to higher priority tasks.
Benefits include:
Critical workloads get fastest response when controller is fully loaded
Storage administrator can make on-the-fly adjustments
Standard feature in Data ONTAP 7.2
FlexShare™
123456789
1
2
Priority mechanism to give preferential treatment to a higher priority tasks
October 23, 2007
FlexVol®
Allows administrator to create multiple flexible volumes across a large pool of disks.
Dynamic, non-disruptive storage (thin) provisioning; space- and time-efficiency.
Allows users to get more space dynamically and non-disruptively.
Enables more productive use of available storage and helps improve performance.
Designed to provide instant replication of data volumes/sets without requiring additional storage space at the time of creation.
Allows IT administrator to make a backup copy of a database and then modify and run testing against test (backup) database without affecting or taking the on-line database off-line.
FlexClone™
Application is free to grab more space if needed
Flexible Volume
October 23, 2007
Designed to survive all 2-disk failure scenarios
Essential for SATA drives but also applicable to FC drives
Breakthrough innovation: USENIX Best Paper in 2004
No performance penalty for industry-leading protection
Automatic failover with notification provides continuous data availability for an unavailable storage controller and notifies administrator of failover.
Manual failover and giveback enables planned maintenance on a storage controller without impacting data availability
Transparent failover: end users are not impacted by a failover; clients continue to access data the same way after a failover.
Active/active configuration; both clustered controllers are doing useful work; no resources sit idle waiting for a failover to occur
Clustered Failover (CFO)
Traditional single-parity RAID technology offers protection from a single failed disk drive.
The expectation is that no other disk fails nor uncorrectable bit errors not occur during a read operation while reconstruction of the failed disk is still in progress.
If either event occurs during reconstruction, then some or all data contained in the RAID array or volume could be lost. With modern larger disk media, the likelihood of an uncorrectable bit error is fairly high, since disk capacities have increased but bit error rates have stayed the same.
The ability of traditional single-parity RAID to protect data is being stretched beyond its limits.
P
DP
Disk
Volumes
October 23, 2007
Instant self-service file & volume backup and recovery for end users
Zero-performance impact, space-efficient. Only block-level changes are stored, thus minimizing disk space for each subsequent Snapshot
Instantaneous volume recovery (up to 16 TB at once)
Can restore individual files/LUNs/volumes as well
Operates within the N series, without burdening the storage networks to which it’s attached
SnapRestore®
October 23, 2007
MultiStore lets you quickly and easily create separate, private logical partitions of a single N series’ IP networking and storage resources.
MultiStore prevents different enterprise departments accessing or finding other virtual storage partitions within the single Appliance or Gateway they all share.
MultiStore helps prevent information on any virtual partition from being viewed, used or downloaded by unauthorized users.
Analogous in purpose to LUN masking on a Block Storage System
SnapMover®
Local data migration solution for optimizing workloads across N series sharing a common disk array.
Allows you to migrate ownership of a volume from one controller to another in the cluster with a single command.
Allows for better resource utilization and performance amongst multiple N series devices
MultiStore and Clustering are prerequisites
Windows clients
Logical partitioning
UNIX clients
Linux clients
N series
B
A
B
A
1
2
1
3
4
2
3
4
October 23, 2007
Async, sync and semi-sync remote replication over inexpensive Internet protocols. RPQ’s for FCP support accepted.
In async mode, a full baseline replication is followed by incremental block updates.
Mirroring can be :
1-to-1
Many-to-1
1-to-Many
SnapMirror®
Synchronous local mirroring from one volume to another volume attached to the same controller.
Includes volume mirroring between two clustered nodes.
Not available on the N3700 due to port limitations
Source
Target
IP
Plex0
Plex1
October 23, 2007
SnapValidator detects and prevents potential corruptions of Oracle data before they happen.
Oracle system assigns a unique value to each write request based on the sum of the data and sends the data set down the I/O path to the N series system. Upon receipt, SnapValidator calculates the sum of the data and compares it to the Oracle value.
If these values match, the system writes the information to disk and reports a successful write.
If the identifiers do not match, the system prevents the data corruption and produces an alert.
SnapValidator meets Oracle HARD (Hardware Assisted Resilient Data) initiative and provides the highest possible level of protection for Oracle data.
SnapValidator®
Extends Clustered Failover capabilities from primary to a remote site
Replicates data from the primary site to the remote site to ensure that data there is completely up-to-date and available
If Site A goes down, MetroCluster allows you to rapidly resume operations at a remote site minutes after a disaster.
Stretch MetroCluster provides a disaster recovery option at distances up to 500 meters between each N series system.
Fabric MetroCluster provides a disaster recovery option at distances up to 100km using a Fibre Channel switched network.
Available on N5000 & N7000 Models
Fabric
X
Ym
Y
Xm
Vol-X
SnapValidator
Fibre Channel
October 23, 2007
NearStore®
Data ONTAP maintains a fixed upper limit for concurrent SnapMirror and SnapVault transfers based on the type of disks the system has attached.
Concurrent operations allow multiple data streams between two N series devices for the purpose of backup, data protection, disaster preparedness
The NearStore (near-line) feature increases the maximum number of concurrent data streams (per storage controller) to the following limits:
N5200 w/FC & SATA drives = 32 concurrent data streams
N5300, 5500 & N7600 w/FC & SATA drives = 64 concurrent data streams
N5600 & N7800 w/FC & SATA drives = 128 concurrent data streams
Notes: Concurrent data streams = a combination of SnapMirror and SnapVault and Open Systems SnapVault sessions
Benefits of A-SIS
Ideal for Business Processes / Applications
Archiving
Between multiple backups of same source data set
Between multiple backups of different source sets
Within one backup (duplicate files and blocks)
Advanced Single Instance Storage (A-SIS deduplicaiton)
Concurrent data streams
October 23, 2007
Provides homogeneous super-efficient hourly disk-based online backup and restore by periodically backing up a Snapshot copy to another system
Full volume, native format copies of primary N series systems reside on backup system and only incremental block changes are sent to backup system for efficient space management
Provides Heterogeneous super-efficient hourly disk-based online backup and restore by periodically backing up changes to an N series system
Full volume, native format copies of data from open system DAS systems reside on backup system but only incremental block changes are stored on backup system for efficient space management
Enables you to provide incremental backup copies from Open Systems with DAS to N series disk systems
Open Systems SnapVault (OSSV)
October 23, 2007
Disk-based WORM technology compliant with strict regulations such as SEC 17a-4. Provides non-erasable and non-rewritable data protection that helps enable compliance with government and industry records retention regulations.
The entire system or a portion of it can be partitioned to store WORM-protected data. SnapLock volumes are accessed via CIFS or NFS protocols. SnapLock works with SnapMirror to satisfy regulations requiring two WORM copies.
Enables IT administrator to “lock” a backup in a non-erasable and non-rewriteable format for compliant retention
Snapshot-based backups save only changed blocks; yet full backup image is immediately readable in native application format
ComplianceJournal™ logs changes between Snapshot copies
LockVault™
October 23, 2007
Host software for managing Exchange and SQL Server backup and restore. SnapManager software simplifies data protection by automating processes to provide hands-off, worry-free data management.
It enables you to schedule and automate backups, use policy-based backup retention management, and simplify the migration of existing databases to IBM N series systems.
SnapManager also delivers built-in high availability, with features that allow you to expand databases online.
It also offers tight integration with Microsoft Cluster Server (MSCS) and Multipath I/O (MPIO) and with the N series Clustered Failover option and SnapMirror for simplified disaster recovery implementation.
Enables the recovery of a single mailbox from a Microsoft Exchange Information Store.
Unlike Lotus Domino where every mailbox is stored as a separate database (and therefore is implicitly recoverable as a unit by itself), Exchange clumps several mailboxes together into .edb files and .stm files.
This makes recovery of a single user’s mailbox extremely cumbersome and in most cases impossible unless there is some brick-level backup software in place.
With SMBR, no brick-level (transaction level) backup is required.
SMBR can extract a single mailbox or email directly and rapidly from an Exchange Information Store.
Single Mailbox Recovery (SMBR)
February 12, 2008
Extraction capability recovers individual Sites, Sub-sites, Libraries, Lists, Files and Versions in minutes
Recovery is non-disruptive
Fast and storage efficient backups
More frequent backups due to speed and efficiency
No performance degradation from online backups
Instantaneous access to backup data
Automation eliminates manual errors
Reduce downtime from outages
Automation saves administrative time
October 23, 2007
SnapManager for Oracle®
SnapManager automates and simplifies the complex manual and time-consuming processes associated with the backup, restore, recovery and cloning of Oracle databases.
It provides the ability to create, use, and clone a database for use by non-production team capability.
SnapManager automatically identifies the backup data set and puts the database in hot backup mode while a Snapshot copy is created to ensure consistency.
Any backup can be immediately verified, or verification can be deferred.
Backups can be performed at regular intervals throughout the day and ensures that restores occur quickly with minimal disruption to ongoing operations.
It also integrates with native Oracle technology such as RAC, RMAN, and ASM and across iSCSI and NFS protocols to allow IT organizations to:
Scale their storage infrastructure
Improve the productivity of databases storage administrator across the enterprise
Benefits:
Snapshot copies verified for integrity and stored
Near instantaneous restores from Snapshot
Dramatically shortened recovery with automated log replays
Automated recovery tasks
October 23, 2007
Optimized usage in Exchange and database environments
Automates many functions that typically require manual processes like creating LUNs and mapping those LUNs to volumes
Virtualized “local” disk to Windows servers. Storage managed by SnapDrive logically appears to come from a locally attached storage subsystem.
SnapDrive allows administrators to easily create virtual disks from pools of storage that can be distributed among several N series Appliances and/or Gateways.
With SnapDrive you add, delete, map, unmap, and mirror virtual disks online. You can expand capacity on-the-fly with no impact to application or system performance
GREEN are LUN’s
October 23, 2007
Previously called: DataFabric Manager, this feature provides the base Operations Manager installation.
Designed to provide a central point of control and provide alerts, reports, and configuration tools.
These tools are designed to help manage storage and content delivery infrastructure, consistent with business requirements, and help maximize availability and reduced total cost of ownership.
Provides advanced function to the Operations Manager Core feature.
Provides detailed storage utilization reports to aid capacity planning, consumption management, data migration, and chargeback.
Operations Manager – Storage Resource Manager (SRM)
.
.
.
.
October 23, 2007
N series disk-to-disk environment.
Effective use of secondary and tertiary storage resources reduces cost
Virtual File Manager™ (VFM®)
Benefits of VFM
Simplified storage administration without client reconfiguration
Reduces recovery time associated with disaster or reconfiguration
Storage consolidation without any client reconfiguration and downtime
Simplifies network storage scalability
Increased productivity through more efficient file retrieval
Links to the data are always maintained
Increased data availability through transparent storage management
Dallas
Virtual File Manager server
N series
The next section contains several solution scenarios
You will need to pick and choose the best solution scenarios that meet your customer’s environment and move the remaining solution slides to the backup section.
You should not need to show all the solution slides to your customers.
Recommended supplement for solutions:
Software Presentations
October 23, 2007
*
provide data protection capabilities
provide Remote Disaster Recovery
store infrequently accessed data
store non-erasable, non-writeable WORM data to meet industry needs
quickly buffer data from production system to cached disk system to tape while dramatically reducing the time it takes to backup production data thus allowing IT managers the ability to meet shrinking backup windows
NearStore (near-line) Storage Solutions
Near-line Storage continues to gain popularity with customers who have a requirement for on-line, fast-access disk solutions
Primary
*
IBM near-line Storage Solution:
Data ONTAP operating system
Populated with SATA disk drive technology
Larger capacity – up to 1,176 TB
Greater data streaming capabilities between primary N series and secondary N series
Runs SnapLock software for non-erasable & non-rewriteable data protection
*
October 23, 2007
*
*
Automated Resource Assurance & DRS
*
VMware and Storage
Higher I/O rates, High Availability, and Disaster Recovery
Cost-effective and space-efficient ‘hot-backups’, restores,
thin-provisioning and cloning
VDI Requires Cost–Effective Storage
Failure to consider storage virtualization in the design and deployment of a VMware Solution simply transfers costs from servers or desktops to storage
October 23, 2007
*
Users access these desktops remotely from a PC or a thin client using a remote display protocol. Since applications are managed centrally at the corporate data center, organizations gain better control over their desktops. Installations, upgrades, patches and backups can be done with more confidence without user intervention.
VMware Virtual Desktop Infrastructure lets companies preserve the security level of their corporate network at offsite facilities. Remote users access applications that reside in the corporate data center, where companies can more easily adhere to compliance requirements as well as secure data that is now locked down in the host country data center.
Joint Value Proposition
Server
Virtual
Machines
Physical Heterogeneous Storage
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
*
Provision in minutes instead of days
Rapid Test and Dev with instant VM cloning
More value
Application level recovery in minutes
Proven best TCO for enterprise storage
Increase storage & server utilization 50-90%
Any protocol - FCP, iSCSI & NFS
October 23, 2007
Network
Virtualized
Servers
Unified
Storage
*
Microsoft Exchange Scenarios
October 23, 2007
*
Reduces implementation time, increase availability and reduce RPO and RTO
Application Servers
Failover to D/R Site
Site or System Disaster
Snapshot
SnapRestore
Automated DB and message store Migration
Optional for MS Exchange:
*
SnapManager for Exchange:
*
Database
Complete in minutes
Recover to any point in time Snapshot™ copy
N series
Time to Backup: Seconds
SnapManager®
SnapDrive™
*
Clicking through the slides, we present a simple scenario…
The simple exchange mailbox servers… Note that they are happily consolidated to a Network Appliance array.
Create a Snapshot – only a couple of seconds to do it! Ready to archive to tape if you want!
SnapDrive and SnapManager – key components. Remember that SnapDrive integrates to the SAN and carries out the requests that SME makes. SME is a wizard driven interface that allows for the automation of Snapshot creation.
Extend the scenario – everything is running famously well, then WHAMMO –
A store is down. SnapRestore to the rescue. SME again is used to handle the work…
SME will apply the appropriate Snapshot, Exchange auto replays the logs, auto validate the store with appropriate Microsoft tools as Microsoft recommends.
Presto – the store is up again and running. And it takes minutes, not hours.
Easy as eatin’ lettuce! See the online Demo for proof! You can easily extend this scenario with SnapVault and send the Snapshots to disk backup on a NearStore.
It may be required to take what the customer knows about a snapshot and re-apply it to how the process of rolling back to it works. Sinking to deep into the technology will obscure the top level message. The detail can come later. Demos help.
Note – the slide says recover in minutes. Compared to tape recover this is very true. Time to recover depends on several factors:
Replay of logs. If the last snapshot is 24 hours old, you will need to run through 24 hours of logs. Obviously it will be far faster than tape, but it is unlikely to be done in 10 minutes. This time can be minimized with more frequent snapshots being taken and kept online. Length of recover is dependent on factors such as CPU and amount of logs to replay.
Time it takes to validate the recover with eseutil.exe.
Processing power of the server
Activity happening on databases in other storage groups on separate LUNS are not affected by the recover scenario.
Transition
There is also one other potential issue you may have – Do you verify your Exchange backups so that you are certain that the consistency of the database is correct so that when a recover occurs? If you do, you may notice the big impact it has on the server CPU.
Exchange with EMC
Cumbersome backup and restore
Centera
CLARiiON CX
CLARiiON CX
*
Under the Hood …
Space cannot be shrunk; growth involves data reshuffling
New spindles are used when volumes expand
No space saving on replication
Exchange Storage Configuration with EMC
Simple Exchange Storage
Space savings on replication
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
*
SnapManager
Simple and scalable platform
SnapLock
SnapLock®
WORM
WORM
Exchange
Exchange
*
Common Procedure
Load the backup to server (entire Exchange database)
Extract the required mailbox
Primary Data Center
Set up recovery server
*
SnapManager backs up Exchange in seconds with snapshots
SMBR restores individual mailboxes from snapshots in minutes
Primary Data Center
October 23, 2007
*
Multiple storage architectures
Single storage architecture
Simplified storage infrastructure
Reduced operational complexity
Simple to scale
IBM N series
Flexible volumes
Greater flexibility with SnapManager for Exchange
Storage leveraged for backup
*
*
Faster recovery from data loss
Improved utilization
October 23, 2007
Secondary
Storage
Tape
*
N5000 w/
FC drives
Storage
Network
IP
*
Let’s start with what we call “Traditional disk based backup and recovery”. On the left you see primary storage which is characterized by a higher cost, higher performance and very fast performance. On the far right you have archive targets which have traditionally been tape or optical jukeboxes. This is characterized by slower performance, often a low cost to store the information but with higher operational costs to manage and access the information. Also, much slower access times to read and write data.
(Click)
2 years ago we introduced the concept of introducing Nearline storage in the middle for disk staging. This enables organizations to do daily backups to disk and backup to tape weekly or bi-weekly which reduces the amount of data that needs to be written to tape. Also, data is online for faster recovery. The other advantage this provides is that you can leverage your existing investment in primary storage, your backup application, and tape libraries.
Automatic failover to mirrored site
Single solution for sync, async, semi-sync mirroring
Solution Benefits
WAN
*
N series DR solution enables an organization to protect it’s critical information in the event of a catastrophic disaster. The N series offerings have changed the economics enabling organizations to not only replicate from primary storage to primary storage but from primary storage to nearline storage so that more data can be replicated than ever before at low cost. Also a single solution supports both sync, semi-sync and async replication. The solution can be installed quickly and requires little in terms of operational costs. It is efficient in that only changed blocks are sent acrosst he network.
SnapMirror Replication – Sync, SemiSync, Async & Bi-directional
Simple and flexible replication solution addresses a broad range of DR needs
Value Proposition
RPO of one minute or less
Dual path option
(sync, async, semi-sync options)
Bandwidth management
Multi-hop, cascading
Re-startable transfers
N series
N series
.
.
*
Disaster Recovery: SnapMirror
Simple and flexible replication solution addresses a broad range of DR needs
Value Proposition
RPO of one minute or less
Dual path option
(sync, async, semi-sync options)
Supported on all N series systems
Bandwidth management
Multi-hop, cascading
Re-startable transfers
October 23, 2007
*
Challenges
October 23, 2007
Workgroup A
Workgroup B
*
High Availability Storage: Move data from slow, unreliable servers with direct attached storage to high performance, highly reliable enterprise-class storage arrays.
Effortless, large scale server consolidation: Consolidate hundreds of file servers with minimal impact on users.
Pooled storage with non-disruptive expansion: Pooled storage for more efficient use of storage resources and non-disruptive expansion for just-in-time provisioning greatly increases your storage utilization levels.
Heterogeneous file sharing: Provide rapid, secure access for data shred to both Unix and Windows users.
Simplified data management: Fewer and simpler storage systems to manage
Seamless integration with existing software and hardware: N series fits seamlessly into your existing environment by embracing open standards and partnering with top data and system management vendors.
Using SnapLock aids compliance archiving
Supports separate retention dates for each archived message
SnapLock is an add-on software license
Leverages existing IBM hardware
Supports RAID-DP for extra data protection
Open, standard data access protocols
Two Versions—SnapLock Compliance & SnapLock enterprise
October 23, 2007
*
SnapLock is non-erasable, non-rewritable magnetic disk storage, also known as WORM (write once, read many).
SnapLock protects archived emails from Intentional or Accidental Erasure or tampering. This is required for some types of compliance (like SEC Rule 17a-4) and strongly recommended for other types of compliance (like Sarbanes-Oxley and HIPAA).
SnapLock supports individual retention dates for each archived message or attachment. The archived messages cannot be deleted until their retention periods have passed. These retention periods can be extended but never shortened.
Because SnapLock leverages existing NetApp hardware and Data ONTAP, it automatically works with almost all popular NetApp features including Snapshots for instant backup, SnapMirror for easy data replication, RAID-DP for a super extra level of protection against disk failures, and the normal high archival and retrieval performance found in NetApp storage.
SnapLock uses standard file protocols to write and read data, so no special API or SDK is required. This helps ensure that data can be read or migrated in the future and protects customers against vendor lock-in.
There are two kinds of SnapLock licenses
SnapLock Compliance: Enables organizations to satisfy strict records-retention regulations such as SEC Rule 17a-4 (broker-dealers), HIPAA (healthcare), Sarbanes-Oxley (public companies), 21CFR Part 11 (life sciences), and DOD 5015.2 (government). Only an act of willful destruction, such as physically removing disks from a NetApp system, can result in record deletion or alteration prior to the specified retention date.
SnapLock Enterprise: Enables adherence to rigorous organizational best practices through functionality similar to that of SnapLock Compliance, but allows administrators to delete entire SnapLock Enterprise volumes. Under no circumstances is it possible for any SnapLock Enterprise user or administrator to delete or modify individual SnapLock Enterprise WORM records or undermine SnapLock Compliance WORM volumes.
“Strict” SnapLock
Trust nobody
Until file expiration
Complies w/ SEC Regulations
Meets SEC 17a-4 requirements
Partial storage admin control
Admin can destroy volumes
Cannot modify/delete individual records
*
Access Controls
LockVaultTM
*
We just discussed LockVault – a data permanence solution for unstructured data that fills a void in the compliant storage market today.
In addition to SnapLock & LockVault, the following features within ONTAP (relating to Data Integrity, Authenticity & Availability) support Data Permanence requirements
SnapMirror®, SyncMirror™, Checksums, RAID DP, RAID Scrubs, MediaScans, Immediate Write Verification
Similarly, a rich set of features within ONTAP help meet Security and Privacy regulatory requirements
Encryption
Secure Deletion
Role Separation
Role-based administration*
Authentication
iSCSI
Set each file’s expiration date
Mark files “read only”
No alterations, overwrites or deletions possible until file expiration (tamper proof ComplianceClockTM)
SnapMirror
Copy WORM protected archive to second site
Retention protection on the archive storage, original date stays with file
Dec 12th 2020
Create SnapLock Volume
October 23, 2007
.
.
*
Key points to highlight –
With SnapLockTM, the process of creating a WORM volume and committing files to it is very simple.
A volume can be created on any existing N series platform – no special purpose box required
File archival (and in fact all file operations) done via open protocols (CIFS/NFS)
Can set retention dates for individual files (as shown here) OR use default retention period on the volume
Transition to read-only commits file to WORM status
SnapLock prevents any tampering with the file until expiration. The retention period itself is monitored using a tamper proof ComplianceClock that can not be spoofed.
After expiration, even on SnapLockTM compliance volumes, expired files can be deleted (with 7.0 release)
Add More Users
Windows & Unix
File Servers
N series Software Bundle
October 23, 2007
.
.
*
Software:
E-mail
Servers
LAN
Windows
*
Alternative Solution
Benefit:
Storage (disks) are consolidated into single poll for better utilization
Reduced complexity and simplified backup solution to a single pool of storage
Solution: Diskless and OS-less Booting
Windows
Servers
.
.
*
Exchange
Servers
.
*
(Click)
Exchange
Servers
LAN
*
(Click)
Solution: SAP Migration & Consolidation
Copies of the SAP production system must be created on a regular base
Setting up a new QA system with current master and transactional data
Testing an SAP upgrade with "real” production data
Typically a backup of the productive database will be restored to the QA system
Long restore time
Not usable for frequent setup of Quality Assurance (QA) systems
Issues:
Database copy can be done using FlexClone/LUN Clone
Created in a few seconds
No double disk space necessary
The clones can be split for long-term usage
Can be automated for frequent setup of QA system
Benefits:
*
Accelerated deployment of CATIA Windows workstations
Clear migration path from V4-V5
V4 filename translation
New projects in V5 with access to legacy V4 files
User-defined table maps CIFS-illegal characters to Windows-recognizable format
During CIFS-list Data ONTAP converts filenames using mapping
Example: file a:b converted to ab for CIFS clients
Performs reverse mapping before actual operation
CIFS request for file ab will always open NFS file a:b
October 23, 2007
WAN
*
N series Deployments
Corporate Data Center
LAN
LAN
Challenges
Automate the process and have a single point of control
Set up/Manage all copy function tasks for single/multiple disk systems
Solutions stack
Value
Eliminate backup window – more frequent backups
Data consistency between both sites
Can bring up remote site with a DB restart
Failover between two active sites
Failover
servers
SQL
Server
Exchange
October 23, 2007
*
Key Point: N series can protect Exchange data from disasters or catastrophic natural events by replication to a remote site with a cost efficient remote mirroring solutions.
Talking Points:
Economical DR –
Any-to-any mirroring– Any N series storage device can mirror to any other storage device- even simultaneously. This provide you the flexibility to mirror to a NearStore for reduced costs or to a high end system for rapid failover.
Bandwidth Efficient – N series replicates only changed data blocks across a mirrored link, dramatically reducing bandwidth requirements for mirrored solutions.
Mirror over existing WAN – N series mirroring solutions work over standard IP based networks
Scalable replication frequency – Adjust replication frequency to replicate data as frequently as every minute or as infrequently as once a day or week - depending on available bandwidth or data protection needs.
Rapid recovery in the event of a disaster: When disaster strikes the primary location, a standby Exchange server at the remote location can connect to the Exchange data on the SnapMirror target volume to provide users with rapid access to their email data.
Challenges
Restores are too slow
Tape recoveries are unreliable
Value
Non-disruptive backup solution offering continuous DB availability – helps meet SLAs
Leverage tiered storage architecture for optimum cost
Low cost database testing/staging
IP or FC SAN
Time to Backup: Seconds
*
Key Point: N series can protect Exchange data from disasters or catastrophic natural events by replication to a remote site with a cost efficient remote mirroring solutions.
Talking Points:
Economical DR –
Any-to-any mirroring– Any Nseries storage device can mirror to any other storage device- even simultaneously. This provide you the flexibility to mirror to a NearStore for reduced costs or to a high end system for rapid failover.
Bandwidth Efficient – N series replicates only changed data blocks across a mirrored link, dramatically reducing bandwidth requirements for mirrored solutions.
Mirror over existing WAN – N series mirroring solutions work over standard IP based networks
Scalable replication frequency – Adjust replication frequency to replicate data as frequently as every minute or as infrequently as once a day or week - depending on available bandwidth or data protection needs.
Rapid recovery in the event of a disaster: When disaster strikes the primary location, a standby Exchange server at the remote location can connect to the Exchange data on the SnapMirror target volume to provide users with rapid access to their email data.
Leverages low cost disk
October 23, 2007
*
Using traditional backup software such as Legato’s Networker or Veritas’ Netbackup, backup data can be staged to a NearStore and subsequently stored to tape for long term archival. In this case backup is done using file level transfers. Backup of laptops using Connected software transfers block level changes.
This solution is not ideal for remote backup because of the amount of data over the network
If your customer is using high end tape drives/libraries and gets higher performance, a single NearStore may not be able to beat the solution in terms of raw performance. However, other benefits exist in terms of reducing the cost and complexity of the backup and restore process.
By performing tape backups from the NearStore, this solution opens up the tape backup window, resulting in better utilization of the tape drives in the library over the longer backup window.
Main tangible benefit of this solution is in terms of backup and restore performance. If backup performance is limited by tape (in the existing environment), NearStore can help speed up backups. Due to the random nature of disk access, restores should be significantly fast (especially single file restore)
Most Tivoli customers already use Disk->Disk->Tape hierarchy, but the intermediary disk is very expensive, so it is only used for short-term staging. NearStore enables keeping far more information readily available on intermediary disk, at very low cost/GB (acquisition and management).
Clearly, if the customer is already using high performance tape drives and performance is a key decision making criteria, NearStore may not be the appropriate solution.
Challenges
Automate the process and have a single point of control
Set up/Manage all copy function tasks for single/multiple disk systems
Solutions stack
Value
Data Consistency between both sites
Can bring up remote site
Windows users
Unix users
*
October 23, 2007
App 2
App 1
App 1
App 3
App 3
App 2
*
As you can see in the before 7G picture, storage is allocated in fixed increments so that unused storage goes to waste. Which means a group that needs more cannot access by any other way than purchasing additional disks.
(build)
With Data ONTAP, a single command can dynamically provision storage so that allocation is optimized, utilization doubles- most customers we talk to are increasing their current storage utilization from 30 & 40% to 70 & 80%- and there’s space available for anyone who needs it.
(build)
So bottom line is- you can use ALL of your disk space!
The technology that makes this possible is called FlexVol, which allows tremendous flexibility to increase or compress the size of volumes to accommodate requirements. And because re-provisioning is automated, you save on the expensive operational costs of taking the system down, paying time and a half on weekends, and the associated business impacts.
To further leverage the capabilities of Data ONTAP, many customers are implementing thin provisioning. Think of how an airline manages it’s booking system by reserving more tickets than seats it has on the plane. With Data ONTAP, you can do the same thing with storage allocation. Several groups may request significant portions of storage but with the flexibility of Data ONTAP, you can overbook or thinly provision their requests.
But don’t take our word for it- here’s one of our customer’s words
Snapshot
SnapRestore
Instantaneous data recovery of volumes up to 16TB or individual files
SnapMirror
FlexClone
Instant and space efficient logical copies providing dramatic space savings
FlexVol
RAID-DP
Data protection against multiple drive failures and Media Bit Errors with nominal impact to application performance
October 23, 2007
*
These are the core technologies of N series which are key to the solutions you’ll see shortly.
I am not going to spend time telling you what I am about to talk about because you all want to get to the real meat of this discussion.
October 23, 2007
File/LUN: my_file.dat
*
Snapshots operate at the block level of the WAFL (Write Anywhere File Layout) file system. As with any type of file system, each file will be comprised of some set of blocks on the physical disk media. WAFL is no different in this respect, and the figure above shows a simplified representation of a file called MYFILE.DAT residing within WAFL.
So what we have here are blocks A, B, and C representing a particular file called myfile.dat with the file system actually sitting above it as a group of pointers to those data blocks.
Consumes no* space!
* Except for 4 KB replicated top-level inode block that defines the Snapshot
How Snapshot works (2)
User’s file (my_file.dat)
October 23, 2007
*
Here we show a Snapshot being taken. A new set of pointers to the blocks of data in the file system is established and all the blocks are made read-only
So, what that means is the snapshot is actually a set of pointers, pointing to the available blocks in the file which are A, B, and C. When a snapshot is taken, the version of MYFILE.DAT that exists within the snapshot is identical to the one in the active file system.
The same set of disk blocks that make up the file in the active file system are used for the Snapshot copy. This is why a snapshot uses almost no additional disk space when it is first created.
This next figure explains why a Snapshot requires very little space as the data continues to change….
*
WAFL file system simple moves the pointers of the Active File to blocks A, B new C’.
Original data is never re-written
Some “copy-on-write” alternatives require three operations (read old data, save old data, overwrite with new) vs N series with a single write.
A
B
C
Active file
Disk blocks
User updates active file and a new block C’ is written with new data.
Old block C is not changed
New data
is written
First Snapshot is composed of blocks A, B & old C
User’s file (my_file.dat)
October 23, 2007
*
In this figure, a network client modifies data at the end of the file, (click) causing the contents of block C to change. As you can see in the picture, the changed block is called C prime.
Let’s look closer at what’s happening here.
WAFL uses a copy-on-write policy that writes the modified block to a new location on disk, creating block C prime.
So as the file changes over time only the changed blocks are recorded, not a new copies of the entire file.
24.bin
*
Active Data of my_file.dat is composed of blocks A, B & New C
Snapshot.1 of my_file.dat is composed of blocks A, B & Old C
Snapshot.2 of my_file.dat is composed of blocks A, B and New ‘C’
Snapshot copies can continue!
New
C
Snapshot.1
A
B
Old C
Active Data
* Finest granularity for a “snapsched” is 1 minute, rolls over after 255 max reached
October 23, 2007
*
Observe in this slide the active data plus two Snapshot copies, yet the overhead is still 33%. Of course, it would be proportionally much smaller if the file/LUN were larger and more of the blocks were common between the active and Snapshot copy.
Now let’s see how a Data ONTAP deals with modified data incrementally.
As you look at the picture on the screen, (click) the active file system now points to blocks A, B and C prime. (click)
The snapshot still points to blocks A, B and C. So, we’ve got a full back-up copy represented by the snapshot and we’ve got another full active file system represented by the blocks pointing to active data. (click) But, the space taken up by these two full copies of data is really just the amount of space in C prime.
OK that’s how a Snapshot works. Let’s explore in more detail a key benefit: End User Driven Recovery.
*
DELETED!
INFECTED!
Keith is free to open this “hidden” folder at any time
User - Keith’s files are stored on Filer8,
which appears to Windows as a “mapped” network drive – just as it would when shared from any Windows server.
“mapped” network drive
October 23, 2007
SAN/NAS
- Oracle
- HomeDirs
1
4
3
2
0
*
The user recovers his own files with a simple copy-paste. This activity necessarily puts to work (1) his workstation and (2) the network connection between his workstation and the storage system. If the amount of data is relatively small (as it would be for a single user), the recovery time would likewise be small.
DELETED!
INFECTED!
Keith may drag any of these read-only files back into active service
October 23, 2007
SAN/NAS
- Oracle
- HomeDirs
1
4
3
2
0
*
Low RTOs (recovery time objectives) can be achieved, but they can hardly be described as near-instantaneous because it takes a finite amount of time for the client to copy the data from the Snapshot copy and write it back. The operation is affected not only by system’s sequential-read and sequential-write performance, but by the interconnecting network speed and speed of the client doing the copy-paste.
I

IBM System Storage™ N series – Unified Storage Solutions © 2007 IBM Corporation An Introduction to IBM System Storage N series Name title October 23,

Documents