This Storage Intelligence (StorInt™) dispatch covers Storage Performance Council (SPC) results. There have been five new SPC-1 and one new SPC-2 submissions since our last report in November. The new SPC-1 results are Huawei OceanStor 6800v3, HDS VSP G1000 (with Hitachi Accelerated Flash), NetApp EF560 all flash array and two from MacroSAN the MS3300 and MS5580. The one new SPC-2 submission was the Fujitsu ETERNUS DX200 S3. Five of our top ten SPC-1 and SPC-2 charts have changed and we present some highlights below.
We begin our discussion with LRT™ (least response time) top ten results shown in Figure 1.
First as discussed in previous reports, all of our top ten LRT systems have a majority of their ASU capacity in SSDs or Flash storage. The new submissions from NetApp and HDS have finally broken the run of IBM/TMS solution in the top 5 in LRT. The NetApp EF560 tied for 2nd place at 180 μsec. and the new HDS VSP G1000 (with Hitachi Accelerated Flash tied for fourth place at 220 μsec. It’s good to see the competition heating up in response time. If anything, recent submissions have been getting slower rather than faster, so the new NetApp EF560 results seems especially noteworthy with its 2nd place ranking.
Next, we turn to Top 10 IOPS, an SPC-1 reported metric, in Figure 2.
In Figure 2, we had to up the scale considerably to accommodate the new number 1, HDS VSP G1000 (with Hitachi Accelerated Flash), with over 2 million IOPS. Also, the new Huawei OceanStor 6800 V3 came in at the 6th position, with ~651 thousand IOPS. Note, the Huawei system was a hybrid solution, with ~1200 600GB disk drives and 160 200 GB SSDs.
As for the HDS VSP G1000, we seldom see this much horsepower in a traditional monolithic storage system but they must be doing something right at Hitachi, because it is blowing every other system completely out of the water here. With the HDS VSP G1000’s #4 LRT and #1 IOPS rankings, it’s going to be pretty hard to beat performance wise.
On the other hand, it’s refreshing to see a non-all flash storage system do well in SPC-1 IOPS. Of the current top 10 IOPS, two are hybrid (Huawei OceanStor 18800 and 6800 v3) and one is disk-only (IBM SVC 6.2 8-node/Storwize V7000).
The last SPC-1 chart will be our IOPs/$/GB rankings in Figure 3.
In Figure 3 we can see the new Huawei OceanStor 6800 V3 came in at 1st place and the new HDS VSP G1000 (with Hitachi Accelerated Flash) came in at 5th place. Hybrid systems still do well on SCI’s IOPs/$/GB, price performance chart as they can still supply relatively good performance for the cost of capacity. As discussed in previous SPC performance reports, there seems to be less of a bias in IOPS/$/GB as compared to the standard SPC-1 $/IOPS metric when it comes to all flash systems.
The new HDS VSP G1000 (with Hitachi Accelerated Flash) seems to be hitting on all the performance metrics we have for SPC-1, as it ranked well in economical performance.
We haven’t reported on SPC-2 metrics recently because of lack of new submissions or new submissions didn’t rank well enough. But the new Fujitsu ETERNUS DX200 S3 has changed that. In Figure 4 one can see the new SPC-2 top ten $/MBPS results.
In Figure 4, Fujitsu’s new ETERNUS DX200 S3 came in at #2 with a $/MBPS of $15.42. The throughput performance of the new DX200 S3, at ~6300 MBPS and its relatively inexpensive price, under $100K, helps it place so well in $/MBPS.
We don’t have a MBPS/$/GB chart for SPC-2 because there aren’t that many all-flash submissions for SPC-2. Unclear why, but sequential IO seems to still perform pretty good on rotating media.
It’s nice to see IO latency receiving some attention from storage vendors. Previously, IBM/TMS seemed to be the only ones trying to advance the art here. But now that NetApp (EF series) and HDS (VSP G1000 series) have arrived, it should get more interesting for everyone.
MacroSAN’s mostly flash MS3300 and MS5580 systems didn’t show on any analysis, as they were a bit longer in response time (0.34 and 0.28 msec.) and a bit short in IOPS (~124K and ~217K) to crack in any of our SPC-1 top ten charts. However, I found their SPC-1 report response time curves quite interesting, as they did worse at 95% load than at 100%, must be some sort of temporary hiccup, suspect garbage collection but can’t be sure. I’m certain we will see better results next time they submit.
I wrote a recent blog post on the problems with testing performance on all-flash storage systems. Essentially, I said we need longer testing to gauge garbage collection, bring into service more realistic block inter-reference patterns and institute some data reduction characterization into the mix so that we could more accurately measure all-flash storage performance. Read the post to learn more.
Any constructive criticisms on how to improve our analyses for any of our performance reports are always welcome. Moreover, if you detect errors in this or any other of our performance reports please do let us know and we will correct it as soon as possible.
[Also we offer more block storage performance information plus our OLTP, Email and Throughput ChampionsCharts™ charts in our recently updated (May 2019) SAN Storage Buying Guide, or for more information on some select ESRP performance results please see our recently updated (May 2019) SAN-NAS Storage Buying Guide, both of which are available for purchase on our website.]
[This performance dispatch was originally sent out to our newsletter subscribers in January of 2014. If you would like to receive this information via email please consider signing up for our free monthly newsletter (see subscription request, above right) and we will send our current issue along with download instructions for this and other reports. Dispatches are posted to our website at least a quarter or more after they are sent to our subscribers, so if you are interested in current results please consider signing up for our newsletter.]
Silverton Consulting, Inc., is a U.S.-based Storage, Strategy & Systems consulting firm offering products and services to the data storage community.