AMD ROCm documentation

Applies to Linux and Windows

2024-11-15

4 min read time

ROCm is an open-source software platform optimized to extract HPC and AI workload performance from AMD Instinct accelerators and AMD Radeon GPUs while maintaining compatibility with industry software frameworks. For more information, see What is ROCm?

If you’re using AMD Radeon™ PRO or Radeon GPUs in a workstation setting with a display connected, review Radeon-specific ROCm documentation.

Installation instructions are available from:

• ROCm installation for Linux

• HIP SDK installation for Windows

• Deep learning frameworks installation

• Build ROCm from source

ROCm documentation is organized into the following categories:

Compatibility

• Compatibility matrix

• Linux system requirements

• Windows system requirements

• Third-party support

• User and kernel-space support matrix

• Docker image support matrix

• Use ROCm on Radeon GPUs

How to

• Use ROCm for AI

• Use ROCm for HPC

• Fine-tune LLMs and inference optimization

• System optimization

• AMD Instinct MI300X performance validation and tuning

• GPU cluster networking

• System debugging

• Use MPI

• Use advanced compiler features

• Set the number of CUs

• ROCm examples

Conceptual

• GPU architecture overview

• GPU memory

• File structure (Linux FHS)

• GPU isolation techniques

• Using CMake

• ROCm & PCIe atomics

• Inception v3 with PyTorch

• Oversubscription of hardware resources

Reference

• ROCm libraries

• ROCm tools, compilers, and runtimes

• Accelerator and GPU hardware specifications

• Precision support