News for hackers. Stuff that matters.

A comment on MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUDA Moat Still Alive. Parent comment

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by mountainriver on Sunday December 22, 2024 @07:58PM (#42490691)
This seems so insane, is anyone actually doing the work to provide an alternative to CUDA? Maybe Google?
[ Parent | on HN ]
Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by ryao on Sunday December 22, 2024 @09:23PM (#42491163)
https://www.intel.com/content/www/us/en/developer/articles/t...
[ Parent | on HN ]
Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by anon291 on Sunday December 22, 2024 @11:30PM (#42491788)
Honestly, probably NVIDIA itself, since they contribute significantly to many open-source projects (MLIR), and also make their SoTA GEMM/Conv implementations open-source and available for study (Cutlass).
[ Parent | on HN ]
Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by david-gpu on Monday December 23, 2024 @06:55AM (#42493770)
*> also make their SoTA GEMM/Conv implementations open-source and available for study (Cutlass)"

Cutlass is a fine piece of engineering, but it is not quite as good as their closed source libraries in real world workloads. There is secret sauce that is not open sourced.

[ Parent | on HN ]
Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by pjmlp on Monday December 23, 2024 @05:44AM (#42493437)
The problem is that everyone that tries, keeps missing the CUDA forest and focus only in a specific kind of tree.
[ Parent | on HN ]