Impact of Mixed--Parallelism on Parallel Implementations of Strassen and Winograd Matrix Multiplication Algorithms - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2002

Impact of Mixed--Parallelism on Parallel Implementations of Strassen and Winograd Matrix Multiplication Algorithms

Résumé

In this paper we study the impact of the simultaneous exploitation of data-- and task--parallelism on Strassen and Winograd matrix multiplication algorithms. We present two mixed--parallel implementations. The former follows the phases of the original algorithms while the latter has been designed as the result of a list scheduling algorithm. We give a theoretical comparison- , in terms of memory usage and execution time, between our algorithms and classical data--parallel implementations. This analysis is corroborated by experiments. Finally we give some hints about an heterogeneous version of our algorithms.

Domaines

Autre [cs.OH]
Fichier principal
Vignette du fichier
RR-4482.pdf (364.24 Ko) Télécharger le fichier

Dates et versions

inria-00072106 , version 1 (23-05-2006)

Identifiants

  • HAL Id : inria-00072106 , version 1

Citer

Frédéric Desprez, Frédéric Suter. Impact of Mixed--Parallelism on Parallel Implementations of Strassen and Winograd Matrix Multiplication Algorithms. [Research Report] RR-4482, INRIA. 2002. ⟨inria-00072106⟩
112 Consultations
310 Téléchargements

Partager

Gmail Facebook X LinkedIn More