AGULLO, Emmanuel; BUTTARI, Alfredo; COULAUD, Olivier; EYRAUD-DUBOIS, Lionel; FAVERGE, Mathieu; FRANC, Alain; GUERMOUCHE, Abdou; JEGO, Antoine; PERESSONI, Romain; PRUVOST, Florent

Métadonnées

Afficher la notice complète

Licence d’utilisation du document

AGULLO, Emmanuel
COmposabilité Numerique and parallèle pour le CAlcul haute performanCE [CONCACE]

BUTTARI, Alfredo
Algorithmes Parallèles et Optimisation [IRIT-APO]
Centre National de la Recherche Scientifique [CNRS]

COULAUD, Olivier
COmposabilité Numerique and parallèle pour le CAlcul haute performanCE [CONCACE]

Langue

Communication dans un congrès

Ce document a été publié dans

International Parallel and Distributed Processing Symposium, International Parallel and Distributed Processing Symposium, IPDPS 2023 - 37th International Parallel and Distributed Processing Symposium, 2023-05-15, St. Petersburg, FL. 2023-06p. 357-367

Résumé en anglais

Dense matrix multiplication involving a symmetric input matrix (SYMM) is implemented in reference distributed-memory codes with the same data distribution as its general analogue (GEMM). We show that, when the symmetric matrix is dominant, such a 2D block-cyclic (2D BC) scheme leads to a lower arithmetic intensity (AI) of SYMM than that of GEMM by a factor of 2. We propose alternative data distributions preserving the memory benefit of SYMM of storing only half of the matrix while achieving up to the same AI as GEMM. We also show that, in the case we can afford the same memory footprint as GEMM, SYMM can achieve a higher AI. We propose a task-based design of SYMM independent of the data distribution. This design allows for scalable A-stationary SYMM with which all discussed data distributions, may they be very irregular, can be easily assessed. We have integrated the resulting code in a reduction dimension algorithm involving a randomized singular value decomposition dominated by SYMM. An experimental study shows a compelling impact on performance.< Réduire

Mots clés en anglais

Matrix multiplication

SYMM

GEMM

2DBC

task-based programming

Symmetric

SBC

TBC

2.5D

Project ANR

Solveurs pour architectures hétérogènes utilisant des supports d'exécution, objectif scalabilité - ANR-19-CE46-0009

Origine

Importé de hal

Unités de recherche

BioGeCo (Biodiversité Gènes & Communautés) - UMR 1202