Cascading GEMM: High Precision from Low Precision

This paper lays out insights and opportunities for implementing higher-precision matrix-matrix multiplication (GEMM) from (in terms of) lower-precision high-performance GEMM. The driving case study approximates double-double precision (FP64x2) GEMM in terms of double precision (FP64) GEMM, leveragin...

Full description

Saved in:
Bibliographic Details
Main Authors Parikh, Devangi N, van de Geijn, Robert A, Henry, Greg M
Format Journal Article
LanguageEnglish
Published 07.03.2023
Subjects
Computer Science - Mathematical Software
Online AccessGet full text
DOI10.48550/arxiv.2303.04353

Cover

More Information
Summary:This paper lays out insights and opportunities for implementing higher-precision matrix-matrix multiplication (GEMM) from (in terms of) lower-precision high-performance GEMM. The driving case study approximates double-double precision (FP64x2) GEMM in terms of double precision (FP64) GEMM, leveraging how the BLAS-like Library Instantiation Software (BLIS) framework refactors the Goto Algorithm. With this, it is shown how approximate FP64x2 GEMM accuracy can be cast in terms of ten ``cascading'' FP64 GEMMs. Promising results from preliminary performance and accuracy experiments are reported. The demonstrated techniques open up new research directions for more general cascading of higher-precision computation in terms of lower-precision computation for GEMM-like functionality.
DOI:10.48550/arxiv.2303.04353