A Look At The AMD EPYC Performance On The Amazon EC2 Cloud

Written by Michael Larabel in Software on 7 November 2018 at 04:51 PM EST. Page 3 of 3. 9 Comments.
Benchmark Result
Benchmark Result

With the Blender 3D renderer in the cloud, the AMD EPYC performance was coming in quite close to the Intel M5 instances. It will certainly be very interesting to see how the 7nm EPYC servers take on Intel Xeon Cascade Lake in 2019.

Benchmark Result

With the Parboil OpenMP MRI Gridding benchmark, the M5A instances were leading over EPYC.

Benchmark Result

While in a computational fluid dynamics test in the cloud via Rodinia, EPYC was trailing the Intel M5 instances except for the basic xlarge instance.

Benchmark Result

With the M-Queens solver for N-Queens, the EPYC performance was on par to slightly faster than the Intel Xeon performance while for the small xlarge size is where the EPYC performance was significantly better than the Xeon Platinum based instance.

Benchmark Result

This also led to rather nice performance-per-dollar for the on-demand EC2 EPYC cloud.

Benchmark Result
Benchmark Result

The BRL-CAD performance was leading on the Intel side for raw performance but when factoring in the 10% lower cost on the EPYC side, the value was close for this solid modeling benchmark.

The real stand-out in the benchmarks run on the Amazon EC2 EPYC instances so far was the m5a.xlarge instance that often tended to deliver superior performance to the Intel-based m5.xlarge instance while costing less. With the larger sizes with both these EPYC 7571 and Xeon Platinum 8175M instances having the same vCPU/ECU count, the Intel performance in several instances came out slightly ahead but with the EPYC-based solutions currently costing 10% less. We'll see if the price difference improves in AMD's favor as more EPYC-based instances roll-put as the Intel hardware costs compared to AMD EPYC are likely much more than 10%, depending upon the discounts Amazon AWS receives from each vendor. In a number of workloads the performance was very close and in a fair share of these tests the M5A instances did come out faster, making us quite anxious for 2019 to see how the 7nm EPYC and Xeon Cascade Lake performance will compete.

With just beginning the Amazon EC2 M5a benchmarks yesterday, these are the initial benchmarks done so far. Stay tuned for some other fresh Intel/AMD cloud Linux benchmarks coming up soon on Phoronix.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.


Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.