...
| Column | ||
|---|---|---|
|
...
|
This page summarises the information needed to start using the Setonix GPU partitions.
| Column | ||||||||
|---|---|---|---|---|---|---|---|---|
|
...
The GPU partition of Setonix is made up of of 192 nodes, 38 of which are high memory nodes (512 GB RAM instead of 256GB). Each GPU node features 4 AMD MI250X GPUs, as depicted in figure Figure 1. Each MI250X comprises 2 Graphics Complex Die (GCD), with each effectively seen as a standalone GPU by the system. A 64-core AMD Trento CPU is connected to the four MI250X with the AMD InfinityFabric interconnect, the same interconnection between the GPU cards, with a peak bandwidth of 200Gb/s—more information at s. For more information refer to the Setonix General Information.
...
| Name | Description |
|---|---|
| rocFFT | Fast Furier Fourier Transform. Documentation pages (external site). |
| rocBLAS | rocBLAS is the AMD library for Basic Linear Algebra Subprograms (BLAS) on the ROCm platform. Documentation pages (external site). |
| rocSOLVER | rocSOLVER is a work-in-progress implementation of a subset of LAPACK functionality on the ROCm platform. Documentation pages (external site). |
...
The main ROCm installation is rocm/5.0.2 provided by HPE Cray. In addition Pawsey staff installed rocm/5.4.3 from source using ROCm-from-source. It is an experimental installation and users might encounter compilation or linking errors. You are encouraged to explore it during development and to report any issues. For production jobs, however, we currently recommend sticking to using rocm/5.0.2.
Submitting Jobs
...
The complete AMD documentation on how to program with HIP can be found here (external site).
Uptake Projects
Uptake projects are collaborations between researchers and Pawsey staff to optimize scientific code for Setonix and advance research, faster. Contact the helpdesk for more information: help@pawsey.org.au.