NFS smackdown: NetApp knocks EMC out
Flash Cache as good as SSD tier
EMC blew NetApp and others away with a CIFS benchmark in January. Now NetApp has returned that favour with interest, blowing EMC away in an NFS benchmark.
The two benchmarks are the CFS and NFS versions of the SPECsfs2008 file access benchmarks. What gives them special piquancy is that this is the first public performance match between two different approaches to flash memory use in storage arrays with the latest controller technology.
EMC has led with idea of using solid state drives (SSD) as a high-performance tier of storage in its VMAX and other arrays. NetApp favours the use of flash as a storage controller cache, its Flash Cache technology. Which is best? Here are are the numbers.
In January EMC's Celerra NS-G8 NAS gateway product front-ended a VMAX with 96 SSDs and scored 118,463 IOPS with an average latency of 1.92 msecs. NetApp's FAS3140 with a Flash Cache could only do 55,398 IOPS with 1.25 msec latency.
Celerra NS-G8 technology achieved 110,621 IOPS and 2.32 msecs latency in the NFS version of the benchmark. A VG8 Celerra with one less data mover engine and a Fibre Channel disk drive VMAX achieved a higher 135,521 IOPS in August this year.
Now NetApp has released benchmark numbers for its new FAS3270 and FAS6240 arrays, the latter with Flash Cache. The 3270 is rated at 101,183 IOPS, much better than the 3140 but failing to beat the Celerra NS-G8 SPECsfs2008 NFS numbers. The 6240 though, achieved 190,675 IOPS, 41 per cent faster than the Celerra VG8, and the second highest SPECsfs2008 NFS score to date.
HP leads the field with a 333,574 IOPS score achieved by a four-node BL860c cluster, using Itanium CPUs, not X86 ones.
What NetApp has done here is to show that, in file-access speed, its Flash Cache-using FAS arrays are superior to the SSD-using VMAX in the configurations tested and with current controller CPU resources. There seems to be no inherent limitation in using a storage controller flash cache instead of an SSD storage array tier. ®
Small correction - 3270 was not using Flash Cache :)
Hi, Dimitris from NetApp here.
The 3270 wasn't using Flash Cache at all, we just wanted to have a result without it.
The 3210 and 6240 both were, and it helps with latency a lot as you can see.
What is also important: All the NetApp configs were a fraction of the price of the EMC ones and provided many times the amount of space.
More details here http://bit.ly/awIYXz and here http://bit.ly/bJZpRD
EMC much more expensive than NetApp
Sent to me and posted anonymously:-
"You missed a critical metric in the comparison - cost per IOP. The EMC lab queen system is likely 10x more expensive than the NetApp kit. For it to be only slightly better in performance at that cost is laughable and you should point it out since your article implies they are similar cost configurations."
Thanks for this ... Chris.
Didn't the benchmarks prove otherwise??
"Your cache becomes a bottleneck, because it isn't large enough to handle the competing storage access demands.
Better to build the whole array out of Flash, and (if necessary) use on-controller RAM for cache, methinks, than to use a flash cache with mechanicals hanging off the tail-end. "
The FlashCache on the NetApp system was only a fraction of the dataset used in the benchmark (~4.5% of the dataset, ~1.2% of the exported capacity), yet it had 172% more IOPs, delivering them with ~half the latency (198% faster). The system also had 488% the exported capacity at a fraction of the cost of the all-SSD system...
As opposed to the 'large sequential' workload, where predictive cache algorithms can really shine, I see only one use case for SSDs:
Unpredictably random read workloads, where the whole dataset fits into the SSD-provided space. I know one such application, NetApps with SSD-shelfs were deployed after extensively testing multiple vendor's solutions.