This course will give a deep dive into the AMD Instinct™ GPU architecture and its ROCm™ ecosystem, including the tools to develop or port HPC or AI applications to AMD GPUs. Participants will be introduced to the programming models for discrete GPUs and APU. This course gives you crucial insights how to use and program HLRS's system Hunter which is equipped with the MI300A APUs and the next system Herder.
The new unified memory programming model makes writing HPC applications much easier for a wide range of GPU programming models. We will cover how to use pragma-based languages such as OpenMP, the basic GPU programming language HIP, and the performance portable language Kokkos. In addition, there will be presentations on other important topics such as GPU-aware MPI, and Affinity. The AMD tool suite, including the debugger and the profiling tools will also be covered.
A short introduction will be given into using GPUs with Python and container workflows, in particular for AI workflows.
Fees
Free of charge.
Pre-required logistics
Some knowledge in GPU and/or HPC programming. Participants should have an application developer's general knowledge of computer hardware, operating systems, and at least one HPC programming language.