Google Cloud recently launched [1]their N4A series powered by their in-house Axion ARM64 processors . In that launch-day benchmarking last month was looking at how the N4A with Axion compared to their prior-generation ARM64 VMs powered by Ampere Altra. There were dramatic generational gains, but how does the N4A stand up to the AMD EPYC and Intel Xeon instances? Here are some follow-up benchmarks I had done to explore the N4A performance against the Intel Xeon N4 and AMD EPYC N4D series.

Today's benchmarking is looking at how the 16 vCPU N4A instance powered by Google Axion processors compares to the 16 vCPU N4 instance powered by Intel Xeon Emerald Rapids and then the 16 vCPU N4D instance powered by AMD EPYC Turin. The same 16 vCPU sizing was used throughout testing all these N4-series VMs and using 400GB of default storage configuration and running Ubuntu 25.10 for a leading-edge Linux software stack.

[2]

The Google N4 instance was using an Intel Xeon Platinum 8581C processor. Note that Google is using Emerald Rapids for N4 rather than the newer Intel Xeon 6P Granite Rapids processors. The Google Cloud N4D was using an AMD EPYC 9B45 processor Zen 5 (Turin) model. Both the N4 and N4D instances arrived at 16 vCPUs by using eight physical cores plus SMT/HT for a combined 16 vCPUs where as the Axion-powered N4A lacks SMT and thus 16 physical cores.

Besides looking at the raw performance in a variety of benchmarks, the performance-per-dollar was also calculated using the hourly pricing for these 16 vCPU instances. With the pricing in Iowa for testing it came down to the N4A instance at $0.71 per hour, the N4D EPYC VM at $0.77 per hour, and the N4 Xeon VM at $0.82 per hour with current pricing. Thanks to Google for having provided free access to the N4A instances during the preview testing.



[1] https://www.phoronix.com/review/google-cloud-n4a-axion

[2] https://www.phoronix.com/image-viewer.php?id=google-cloud-n4-arm64-epyc-xeon&image=google_cloud_n4_2_lrg