Versions Compared

Old Version 24

changes.mady.by.user Maciej Cytowski

Saved on

compared with

New Version 25

changes.mady.by.user Alexis Espinosa

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.


The Pawsey Partner Scheme supports the Pawsey partner institutions: CSIRO, Curtin University, Murdoch University and the University of Western Australia. The scheme provides significant amounts of compute time for meritorious projects led by researchers from these institutions. 

...

Info
titleHow to estimate Service Units request for Setonix-GPU?

Researchers planning their migration from NVIDIA-based GPU systems like Pawsey’s Topaz and NCI’s Gadi to AMD-based Setonix-GPU should use the following example strategy to calculate their Service Units request.   

  • Simulation walltime on a single NVIDIA V100 GPU: 1h
  • Safe estimate for Service Units usage on a single Setonix’s AMD MI250X GCD: 1h * 1/2 * 128 = 64 Service Units (note: each AMD MI250X GPU has 2 GCDs for the total of 8 GCDs per node)

Please see: https://www.amd.com/en/graphics/server-accelerators-benchmarks 

...

Info
titleSetonix-GPU migration pathway

The Setonix’s AMD MI250X GPUs have a very specific migration pathway related to CUDA to HIP and OpenACC to OpenMP conversions. Pawsey is working closely with research groups within PaCER project (https://pawsey.org.au/pacer/) and with vendors to further extend the list of supported codes. 

Please see: https://www.amd.com/en/technologies/infinity-hub  

...