DGEMM on Integer Matrix Multiplication Unit · HackerPapers discussions

DGEMM on Integer Matrix Multiplication Unit — mentioned by touisteur

Mon, 24 Mar 2025 19:18:01 GMT

I'm particularly fond of the Ozaki scheme https://arxiv.org/html/2306.11975v4 and its recent refinements. Hopefully it trickles down to standard HPC libraries soon.

Paper and discussion history. A mention is not an endorsement.

DGEMM on Integer Matrix Multiplication Unit — mentioned by touisteur

Wed, 09 Oct 2024 22:16:26 GMT

Ozaki has been doing fp64 matrix-multiplication using int8 tensor cores https://arxiv.org/html/2306.11975v4 Interesting AF.

Paper and discussion history. A mention is not an endorsement.

DGEMM on Integer Matrix Multiplication Unit — mentioned by touisteur

Thu, 18 Jul 2024 18:54:11 GMT

Not really any paradigm shift since the introduction of Tensor Cores in NVIDIA archs. Anything Ampere or Lovelace, will do to teach yourself CUDA up to the crazy optimization techniques and the worst libraries that warp the mind. You'll only miss on HBM which allows you to cheat on memory bandwidth, amount of VRAM (teach yourself on smaller models...), double precision perf and double precision tensor cores (go for an A30 then and not sure they'll keep them - either the x30 bin, or DP tensor cores - ever since "DGEMM on Integer Matrix Multiplication Unit" - https://arxiv.org/html/2306.11975v4…

Paper and discussion history. A mention is not an endorsement.