SCI’s latest SPC performance report as of February’17

In $/IOPS, $/MBPS, A700S, AccelStor NeoSapphire 3602, EF560, EonStor DS 4024B, ETERNUS AF650, IOPS, LRT, OceanStor 18500, SPC, SPC-1, SPC-2, StoreServ 20850 AFA by AdministratorLeave a Comment

This Storage Intelligence (StorInt™) dispatch covers Storage Performance Council (SPC) results[1]. There have been two new (AccelStor NeoSapphire 3602 and Huawei OceanStor™ 2600 v3) SPC-1 V1 submissions, two new (NetApp AFF A700s and Fujitsu ETERNUS AF650) SPC-1 V2 submissions and two new (Huawei OceanStor 18500 v3 and 6800 SPC-1 V3 submissions since our last report. For now, we will include all V3, V2 and V1 SPC-1 submissions together in our analysis. The major IO difference between the V1 and V3, is that V3 uses a larger fixed block size than V1, (8KB vs. 4KB)[2] and while this might take an extra microsecond or two to transfer at 16Gbps, it seems unlikely to impact IOPS and LRT results that much. I haven’t investigated the differences between V2 and V1 but SPC reports these results together. There has only been one new SPC-2 submission, the Infortrend EonStor DS 4024B.

SPC-1 results

We begin our discussion with top ten SPC-1 IOPS™ (IO/sec.) results shown in Figure 1.

Figure 1 Top 10 SPC-1 IOPS results

In Figure 1 we can see the NetApp AFF A700s and Huawei OceanStor 18500 v3 came in 3rd and 4th with 2.4M and 2.3M IOPS, respectively. The NetApp A700s used 144 960GB SSDs across 6 pairs of HA controllers with 6.1TB of DRAM while the Huawei 18500 v3 used 240 900GB SSDs across 16 pairs of active-active controllers with 4TB of DRAM. Unclear to me whether having more SSDs in a configuration helps as much as having more disks but the #1 DataCore with over 5.1M IOPS only used 54 300GB SSDs.

Next, we turn to SPC-1 top ten LRT™ in Figure 2.

Figure 2 SPC-1 Top 10 LRT™

In Figure 2 the only new submission was the 7th place, AccelStor NeoSapphire 3602 with an LRT of 0.21 ms. AccelStor is a ~3-year-old storage company that comes out of Taiwan and have a full family of AFA storage solutions. Apparently, they know how to design a system to provide good IO response times. The AccelStor 3602 used 10 480 GB SSDs with only 128GB of DRAM generating an impressive LRT of just over 0.2ms for ~60K IO/s (10% of their max of ~600K IOPS).

We turn now to Top $/IOPS for SPC-1 in Figure 3.

Figure 3 Top 10 SPC-1 $/IOPS

In Figure 3, once again the only new submission was the AccelStor 3602 coming in 3rd place with an $/IOPS of under $0.20/IOPS. This speaks to the fact that the AccelStor 3602 only costs around $40K and delivered over 600K IOPS, again an impressive showing for a relative newcomer.

SPC-2 results

Next, we turn to new SPC-2 results with Figure 4 that shows the Top $/MBPS SPC-2 results.

Figure 4 Top SPC-2 $/MBPS results

In Figure 4 we see that the new Infortrend EonStor DS 4024B came in first place with an $/MBPS of $6.80. The Infortrend DS 4024B is an all flash system that used 20 400GB SSDs.

AFA or majority SSD submissions for SPC-2 are still rare. There were only 2 other top 10 $/MBPS systems that were AFA, the (#2) NetApp EF560 AFA and the (#9) HP 3PAR StoreServ 20850. The Infortrend DS 4024B had a purchase price of under ~$70K and supported an MBPS rate of over ~10GB/sec, which provided the SPC-2 price performance of $6.80/MBPS. 


Unclear how long SPC will continue to allow V1 and V2 new submissions since they have released V3. Combining all of them into the same top 10 like we do above is not kosher according to SPC but we feel they are all similar enough that doing so provides a valid comparison. We reserve the right to split V3 from V1-V2 submissions in the future if we become convinced that the V3 and V1-V2 workloads are different enough to warrant unique treatment.

Seeing a (relatively) new AFA storage system vendor do well in SPC-1 results is always good news for the health of the industry. The new AccelStor seems very interesting and we look forward to seeing more top performance results in the future from this company. Many other AFA startups use compression/deduplication which makes them unsuitable for SPC-1 benchmarks and we seldom see them compete here because of that. But clearly at least some storage startups are providing excellent storage performance even though they are under-represented here.

As always, suggestions on how to improve any of our performance analyses are welcomed.  Additionally, if you are interested in more block storage performance information, checkout our recently updated (February 2017), SAN Storage Buying Guide available for purchase from our website[3].  In the SAN Storage Buying Guide we now provide top 30 results for some of our SPC performance charts and new, SCI derived, OLTP, Throughput and Email ChampionsCharts™ for enterprise, all-flash, mid-range and SMB class SAN storage systems.

[Also we offer more block storage performance information plus our OLTP, Email and Throughput ChampionsCharts™ in our recently updated (May 2019) SAN Storage Buying Guide, or for more information on protocol 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 February of 2017.  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

[1] All SPC results available from as of 24Nov2016

[2] SPC-1 V3 specification can be found at

[3] Available at

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.