Intel Xeon Max 9480/9468 Show Significant Uplift In HPC & AI Workloads With HBM2e

Written by Michael Larabel in Processors on 28 June 2023 at 05:00 PM EDT. Page 6 of 7. 26 Comments.
OpenVINO benchmark with settings of Model: Person Detection FP16, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Person Detection FP16, Device: CPU. Xeon Max 9468 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Person Detection FP32, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Person Detection FP32, Device: CPU. Xeon Max 9468 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Face Detection FP16-INT8, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Face Detection FP16-INT8, Device: CPU. Xeon Max 9468 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Weld Porosity Detection FP16, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Weld Porosity Detection FP16-INT8, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.
OpenVINO benchmark with settings of Model: Person Vehicle Bike Detection FP16, Device: CPU. Xeon Max 9480 2P: HBM Only was the fastest.

The performance results with Intel Xeon Max were great out of Intel's open-source OpenVINO toolkit. With OpenVINO able to leverage the Advanced Matrix Extensions (AMX) and benefit significantly from HBM2E-only operation in the benchmarks, there were many significant speed-ups seen across a variety of tested models.

PETSc benchmark with settings of Test: Streams. Xeon Max 9468 2P: HBM Only was the fastest.
Stress-NG benchmark with settings of Test: Matrix 3D Math. Xeon Max 9468 2P: HBM Only was the fastest.
nekRS benchmark with settings of Input: TurboPipe Periodic. Xeon Max 9468 2P: HBM Only was the fastest.

The PetSC library's streams performance saw huge uplift with the HBM2E memory.


Related Articles