Sign in to my dashboard Create an account

Keep on Scalin’: How FlexGroup Volumes Scale Out Performance

Justin Parisi

NetApp has a bit of a compulsion. We keep publishing SPEC SFS®2014_swbuild numbers! And why wouldn’t we, when we keep hitting #1 on the list?

But, right now, we’re in a bit of an arms race against another vendor that also loves to post SPEC SFS®2014_swbuild numbers. After we posted #1 numbers in November, they have since posted new numbers and were able to surpass our 8-node submission of 4200 builds. They hit 5700 builds (though they needed 23 nodes and 46 100Gbe interfaces to do it).

What’s interesting is that NetApp’s submissions always leave storage efficiencies enabled, providing the most bang for your capacity buck. I don’t see any mention of storage efficiencies enabled in the competitor’s newest submission. We keep ours enabled because we can – ONTAP performance barely suffers when we keep all of our space saving features on.

We also didn’t modify any of the client-side attribute cache settings. Everything was left as the default values – this shows that we can achieve high performance without taking any shortcuts and increasing the overall simplicity story.

But all that is moot – NetApp has upped the ante and submitted *another* SPEC SFS®2014_swbuild publication – this time, with 12 nodes! And, guess who’s #1 again at 6200 sustained builds?


Our new 12-node NetApp ONTAP 9.5 cluster run of the SPEC SFS®2014_swbuild benchmark resulted in 6200 builds with an ORT of .83 ms. So, with FlexGroup volumes, NetApp was able to outperform every vendor that submitted SPEC SFS®2014_swbuild results, with half the nodes needed and much less network gear involved.

NetApp also showed that FlexGroup volumes can deliver on scale – both for capacity and performance. In a real world workload, where you’ve hit the limits of the system, you can simply add new nodes and disks and non-disruptively grow the FlexGroup across the new nodes. No outages. No maintenance window needed.

FlexGroup volumes are great for a variety of workloads, but especially for workloads such as engineering and design applications (EDA), software repositories, AI and machine learning, data lakes and many more. In fact, NetApp Active IQ uses FlexGroup volumes today for their data lake needs.

The Performance Results

As you would expect, adding more nodes to a cluster brings additional performance. The following graph shows the past three SPEC SFS®2014_swbuild submissions from NetApp running ONTAP 9.5. The new 12-node cluster submission extends into the 3 million Ops/Sec range while maintaining sub-3ms latency over NFSv3. That’s near-linear scale of 3x the Ops/Sec as compared to the 4-node cluster submission.


This chart also shows how an ONTAP cluster was able to scale build numbers linearly as nodes were added.


NetApp has proven yet again that ONTAP with FlexGroup volumes can do more work at lower latencies and higher throughput for faster overall completion times. And we all know time is money, so expect return on investment over the life of the system as well.

These combined results show that with ONTAP and NetApp FlexGroup volumes, you don’t have to decide between IOPS, latency, throughput, or the number of builds a cluster can handle. You get it all. In addition to performance, you also get:

  • Cost benefits of storage efficiencies
  • FabricPool support (AFF only)
  • Nondisruptive scale-out and scale-up of file systems and clusters to grow with your applications
  • Enterprise-class Snapshot™ technology
  • Simple, streamlined deployment

To learn more, check out:

Justin Parisi

Justin Parisi is an ONTAP veteran, with over 10 years of experience at NetApp. He is co-host of the NetApp Tech ONTAP Podcast, Technical Advisor with the NetApp A-Team and works as a Technical Marketing Engineer at NetApp, focusing on FlexGroup volumes and all things NAS-related.

View all Posts by Justin Parisi

Next Steps

Drift chat loading