AMD EPYC 9755 DDR5-4800 vs. DDR5-6000 Memory Performance

Written by Michael Larabel in Memory on 14 October 2024 at 10:45 AM EDT. Page 2 of 9. 3 Comments.
Timed Linux Kernel Compilation benchmark with settings of Build: defconfig. DDR5-6000 was the fastest.
Timed Linux Kernel Compilation benchmark with settings of Build: allmodconfig. DDR5-4800 was the fastest.
Timed Godot Game Engine Compilation benchmark with settings of Time To Compile. DDR5-6000 was the fastest.
Timed Node.js Compilation benchmark with settings of Time To Compile. DDR5-6000 was the fastest.
Timed Gem5 Compilation benchmark with settings of Time To Compile. DDR5-6000 was the fastest.
Timed LLVM Compilation benchmark with settings of Build System: Ninja. DDR5-6000 was the fastest.

For those considering AMD EPYC 9005 series for a CI/CD build box type deployment just doing a lot of code compilation, there was minimal benefit to DDR5-6000 over DDR5-4800 speeds. Not entirely surprising but here are the numbers for those interested in trying to weight the performance impact if sticking to DDR4-4800 from a Genoa server upgrade or trying to lower server costs by using the slower memory.

OpenSSL benchmark with settings of Algorithm: SHA256. DDR5-4800 was the fastest.
OpenSSL benchmark with settings of Algorithm: AES-256-GCM. DDR5-4800 was the fastest.
OpenSSL benchmark with settings of Algorithm: ChaCha20-Poly1305. DDR5-4800 was the fastest.
John The Ripper benchmark with settings of Test: WPA PSK. DDR5-6000 was the fastest.
Apache IoTDB benchmark with settings of Device Count: 500, Batch Size Per Write: 100, Sensor Count: 800, Client Number: 400. DDR5-4800 was the fastest.
Apache IoTDB benchmark with settings of Device Count: 800, Batch Size Per Write: 100, Sensor Count: 800, Client Number: 100. DDR5-6000 was the fastest.

For workloads not very memory bandwidth constrained, running the AMD EPYC 9755 with DDR5-4800 was still delivering the same great performance.

Related Articles