Pierre Winiger Technical Partner Manager Product News
Mar 29, 2015
Pierre Winiger
Technical Partner Manager
Product News
New FAS-Systems
Entry-Level Roadmap
FAS2040
FAS2020
FAS3210
Today 3Q 2011 4Q 2011 1Q 2012 2Q 2012
2USAS
4USATA
FAS2240
FAS2040
432TB
144 Drives
408TB
136 Drives
4,320TB
1,440 Drives
6TB Flash Cache
4,320TB
1,440 Drives
8TB Flash Cache
1,800TB
600 Drives
1TB
Flash Cache
3,600TB
1,200 Drives
3TB Flash Cache2,880TB
960 Drives
2TB Flash Cache720TB
240 Drives
512GB
Flash Cache
New Entry Systems Added to Line
Truly unified Most efficient Extremely flexible
FAS/V3210
FAS2040
FAS/V6240
FAS/V6280
FAS/V3240
FAS/V6210
FAS/V3270
FAS2240
Unified Storage Architecture
FAS2240-2 Front View
FAS2240-2 Rear View
FAS2240-4 Front View
FAS2240-4 Rear View
Processor controller modules
Power Supply/Fan Module
FAS2240 I/O ConnectionsWith Mezzanine Card Option
Optional
FAS2240 Configuration Details
Based on SAS shelves– FAS2240-2 uses DS2246 and internal SAS disks
– FAS2240-4 uses DS4243 and internal SATA disks
– Up to 24 internal disks for both form factors
Same controller PCM used in both models4x GbE ports and 2x 6Gb SAS ports per PCMOptional mezzanine card per PCM
– 2x 10GbE - or -
– 2x 8/4/2 Gb FC (target or initiator mode) - or -
– No card, slot can be left empty
FAS2240 Configuration Details
FAS2240-2 internal disks– 12 and 24 disk orderable configurations– 450GB or 600GB 10K SAS supported at launch
FAS2240-4 internal disks– 12 and 24 disk orderable options – 1TB, 2TB SATA, and 3TB SATA available at launch
FAS2240 Back-end SAS Storage
Supported shelves and modules– DS2246 with IOM6– DS4243 with IOM3– Mixing DS2246 and DS4243 supported
Only applies to FAS2240, not other platforms
Supported disks– All SAS disks supported in above shelves– SSDs are supported externally in DS4243 shelves
Available starting January 2012
Set FAS2240 shelf ID with external SAS storage
FAS2240 Back-end FC Storage
Supported shelves and modules– DS14mk4 with ESH4– DS14mk2 with ESH4
ESH2 is not supported in Data ONTAP® 8.1
– DS14mk2-AT with AT-FCXSupported disks
– 144GB and larger 15K FC disks– 500GB and larger SATA disks
Unsupported disks– Any 10K FC disks– 36GB and 72GB 15K FC disks
FAS2240 Upgrade Guidelines
FAS2240 can be converted to storage shelf– Replace PCMs with correct IOMs – Connect to controller– Assign disk ownership
Existing SAS shelves cannot be converted into FAS2240– FAS2240 always ships as a new system– FAS2240 cannot be created in field by
converting a shelf
FAS2240 Software Bundles Data ONTAP® Essentials included with FAS2240
– Includes ALL protocols at $0 charge
– Performance optimization: FlexShare®
– Storage efficiency: Thin provisioning, deduplication, compression, Snapshot™
– Core management: System Manager, Operations Manager, Provisioning Manager, Protection Manager
– High availability: Cluster failover, RAID-DP®, Snapshot, DSM/MPIO, SyncMirror®, OSSV
– Secure multi-tenancy: MultiStore®
Com
ple
te B
undl
eSnapManager® Suite
+ All Protocols
SnapVault®
FlexClone®
SnapMirror®
SnapRestore®
Optional Bundles
FAS2240 and FAS2040 ComparisonHA Configurations
FAS2040 HA FAS2240 HAProcessor Cores 4x 32-bit 4x 64-bit
Hyperthreading CPU Physical Memory 8GB 12GB
Onboard I/O4x 4Gb FC
8x GbE2x 3Gb SAS
8x GbE4x 6Gb SAS
I/O Expansion Option
Optional 10GbE or 8Gb FC Maximum Spindles 136 144
Management BMC Service Processor (SP)
Data ONTAP® Support 7.3.2+, 8.0.x, 8.1.x 8.1
FAS2240 Performance OverviewSFS2008 NFS
FAS2040 FAS2240 FAS3210 FAS3160 FAS3170 FAS3240 FAS32700
5000
10000
15000
20000
25000
30000
35000
40000
20300
38000
SFS2008 NFS Filesystem Performance (HA Pair)7-Mode
Th
rou
gh
pu
t (I
OP
s)
FAS2240 Performance OverviewOLTP
FAS2040 FAS2240 FAS32100
5000
10000
15000
20000
25000
30000
35000
40000
12600
33900
OLTP Database IOPs (HA Pair)7-Mode
Th
rou
gh
pu
t (I
OP
s)
Data ONTAP 8.1
Data ONTAP 8.1 New Feature Overview
Platform support is the same as for 8.0.x plus new FAS 2240
32-bit aggregate expansion and conversion
Larger 64-bit aggregate sizes
Default Snapshot reserve settings reduced– Aggregate snapshot default = 0% (except when using SyncMirror)– Volume snapshot default = 5%
Compelling storage efficiency enhancements– Deduplication, data compression and cloning
Data ONTAP 8.1 New Feature Overview
Non-disruptive 1-step Upgrade Feature parity with Data ONTAP 7.3.3 and later
– SnapLock
DataMotion for vFiler use case expansion– vFiler migration from faster to slower controllers and from
faster to slower storage media is supported
– Enables non-disruptive ‘down-tiering’ of NFS and iSCSI data
Data ONTAP 8.1 New Feature Overview
All new aggregates will default to 64-bit format– Root aggregate will default to 64-bit format
In-place expansion of existing 32-bit aggregates to the 64-bit format
Physical replication between volumes within different aggregate types
32 64 3264
64-bit Aggregate and Volume Size Limits
Maximum aggregate sizes increase for most systems with 8.1 (both modes) Volume limits remain the same; now full size can be achieved 8.1 volume limits also apply with deduplication & compression enabled
FAS/V ModelAggr & Vol Limit 8.0.x
Dedupe Vol Limit 8.0.x
Aggr Limit 8.1
Vol Limit 8.1
6280, 6240 6080, 6070
100 TB 16TB 162 TB 100 TB
6210 70 TB 16TB 162 TB 70 TB
6040, 60303270, 3170
70 TB 16TB 105 TB 70 TB
3240, 3160 50 TB 16TB 90 TB 50 TB
3210, 3140 50 TB 16TB 75 TB 50 TB
3070, 3040 50 TB 16TB 50 TB 50 TB
2040 30 TB 16TB 50 TB 30 TB
In-place Expansion Overview
Availability– Both 7-mode and C-mode supported
– No license required
– Supported on all platforms
Expansion process– Triggered automatically when aggregate is grown beyond 16TB
– Expands the aggregate and all volumes within the aggregate
– The expansion is in-place, non-disruptive and does not require a data copy
In-place Expansion Overview
Performance Impact– Minimal impact on system throughput
– No interruption to storage services during the expansion process
Space usage– On-box tool available to estimate space required for expansion
1 2 3 4 5 6 7 8 90
1
2
3
4
5
6
Baseline:sys_avg_latency
64-bit expansion:sys_avg_latency
1 2 3 4 5 6 7 8 90
0.51
1.52
2.53
3.54
4.55
Baseline:sys_avg_latency
64-bit expansion:sys_avg_latency
SFS benchmark SPC-1 benchmark
Sample completion timesAggregate size Volume size # of
volumes Usage in
each volume
Completion time
16TB 10G 499 50% ~30hrs
16TB 13TB 1 100% ~28hrs
13TB 12TB 1 30% ~10hrs
13TB 1TB 10 90% ~1hr
13TB 100GB 100 70% ~1hr
13TB 25G 499 66% ~1hr
13TB 400G 40 30% ~1hr
Flash Cache Fast Rewarm
Cache data will preserve in FlashCache– For graceful failures
system halt reboot cf takeover
– Only works with FlashCache (no PAM I support)
No rewarming for non-graceful events– panic
– power cycle
– cf forcetakeover
Automatically on, no need to tweak
SAS Tape Drive Support
Starting with Data ONTAP 8.1, SAS tape drives are supported
Has to be connected directly to a SAS port
Deduplication Enhancements
no license keys any moregeneral performance enhancementswill be enabled by default for new FlexVolsDedup continues after HA failuresno maximum dedup limits any more (equals
maximum volume limit for platform)
Compression Enhancements
Is GA now – no PVR required No license key required Inline & Post-Processing (both or post processing only) Integrated with Deduplication:
- same schedule- New options to dedupe CLI commands
Less limits:- no volume size limits - but still for 64bit aggregates only
Support by System Manager 2.0
OnCommand System Manager 2.0Newest Features: Easier, More Powerful than Before
Management of 7-Mode and clusters
Easy wizard setup and configuration: CIFS, NFS, iSCSI, FC, and FCoE
Simple device management: – Intuitive browser– Dashboard view and graphical reports– Automated workflows
Powerful Data ONTAP features:– Storage efficiency and security– High availability and replication– Virtualization
Platforms:– Windows or Linux– NetApp Data ONTAP software– VMware® ESX™ integration wizard
Record Performance, Linear Scaling on Demand
World-Record NFS Performance
NetApp® FAS6240 systems in 24-node cluster Better performance for demanding NAS workloads Higher efficiency saves infrastructure costs Enterprise-class capabilities for nondisruptive
operations
0
500000
1000000
1500000
Top SPECsfs2008 NFS PerformanceNetApp
Isilon
EMCVNX VG8
IBMSONAS
Linear Scaling on Demand
FAS6240 systems running Data ONTAP® Cluster-Mode Start with the right size cluster for today’s storage requirements Nondisruptive, massive unified scalability to handle future needs Consistently fast response time for demanding applications
4 8 12 16 20 24
Ops/sec 260,388 512,667 765,667 1,007,052 1,261,145 1,512,784
ORT 1.53 1.54 1.55 1.48 1.56 1.53
0
0.5
1
1.5
2
2.5
3
3.5
0
200,000
400,000
600,000
800,000
1,000,000
1,200,000
1,400,000
1,600,000
SPEC
sfs2
008
Ove
rall
Res
po
nse
Tim
e (O
RT)
in m
s
SPEC
sfs2
008
nfs
v3 O
ps/
sec
Number of nodes in cluster
NetApp SPECsfs2008 Performance Scaling
Consistent Response Time
Linear Scaling
Leading Performance and Efficiencyfor Demanding NAS Storage Infrastructure
NetApp delivers better value 35% higher performance 40% faster response time 50% fewer disks 80% fewer controller nodes
NetApp Isilon
Ops Per Second (SPECsfs2008 NFS)
1.51M 1.11M
Response Time 1.53ms 2.54ms
# Disks 1728 3360
# Controller Nodes 24 140
Performance
Efficiency
FAS6240 24-node cluster with Data ONTAP® 8