Book covers modern GPU programming techniques for ML
An online resource based on a mini-series in CMU’s ML Systems course covers data layout swizzling, 3D TMA, and programming for Blackwell.
The online book is based on a mini-series taught in CMU’s ML Systems course. It focuses on GPU programming for machine learning systems.
Topics listed include data layout swizzling, 3D TMA, and programming for Blackwell. The material gives engineers course content for exploring these techniques.
As you study, note each concept and connect it to the systems problem you want to understand. Distinguish claims made in the material from tests or conclusions you develop yourself.
If you use AI to summarize or apply the material, avoid including personal data, credentials, or internal organizational information. Use synthetic examples and share only the context needed.