Ubuntu 24.04 + Linux 6.9 Intel & AMD Server Performance

Written by Michael Larabel in Processors on 23 May 2024 at 12:30 PM EDT. Page 8 of 10. 5 Comments.
TensorFlow benchmark with settings of Device: CPU, Batch Size: 256, Model: ResNet-50. Xeon Platinum 8490H 2P was the fastest.
TensorFlow benchmark with settings of Device: CPU, Batch Size: 256, Model: AlexNet. Xeon Platinum 8490H 2P was the fastest.
TensorFlow benchmark with settings of Device: CPU, Batch Size: 512, Model: ResNet-50. Xeon Platinum 8490H 2P was the fastest.
TensorFlow benchmark with settings of Device: CPU, Batch Size: 512, Model: AlexNet. Xeon Platinum 8592+ 2P was the fastest.

The Intel Xeon Emerald Rapids and Sapphire Rapids processors led in TensorFlow thanks to Advanced Matrix Extensions (AMX).

PyTorch benchmark with settings of Device: CPU, Batch Size: 256, Model: ResNet-50. Xeon Platinum 8592+ was the fastest.
PyTorch benchmark with settings of Device: CPU, Batch Size: 512, Model: ResNet-50. Xeon Platinum 8592+ was the fastest.
PyTorch benchmark with settings of Device: CPU, Batch Size: 512, Model: ResNet-152. Xeon Platinum 8592+ was the fastest.

Similarly, the Intel Xeon Scalable CPUs were dominating in PyTorch.

PyTorch benchmark with settings of Device: CPU, Batch Size: 512, Model: ResNet-152. Xeon Platinum 8592+ was the fastest.
PyTorch benchmark with settings of Device: CPU, Batch Size: 512, Model: ResNet-152. Xeon Platinum 8592+ was the fastest.

On a performance-per-Watt basis in PyTorch, the AMD EPYC 8004 Siena processors were delivering the best performance-per-Watt.


Related Articles