Data Distribution Schemes for Dense Linear Algebra Factorizations on Any Number of Nodes - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Data Distribution Schemes for Dense Linear Algebra Factorizations on Any Number of Nodes

Quelques schemas de distribution de données en algèbre linéaire pour un nombre quelconque de noeuds

Abstract

In this paper, we consider the problem of distributing the tiles of a dense matrix onto a set of homogeneous nodes. We consider both the case of non-symmetric (LU) and symmetric (Cholesky) factorizations. The efficiency of the well-known 2D Block-Cyclic (2DBC) distribution degrades significantly if the number of nodes P cannot be written as the product of two close numbers. Similarly, the recently introduced Symmetric Block Cyclic (SBC) distribution is only valid for specific values of P. In both contexts, we propose generalizations of these distributions to adapt them to any number of nodes. We show that this provides improvements to existing schemes (2DBC and SBC) both in theory and in practice, using the flexibility and ease of programming induced by task-based runtime systems like Chameleon and StarPU.
Fichier principal
Vignette du fichier
article.pdf (356.62 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04013708 , version 1 (03-03-2023)

Licence

Attribution

Identifiers

  • HAL Id : hal-04013708 , version 1

Cite

Olivier Beaumont, Jean-Alexandre Collin, Lionel Eyraud-Dubois, Mathieu Vérité. Data Distribution Schemes for Dense Linear Algebra Factorizations on Any Number of Nodes. IPDPS 2023 - 37th IEEE International Parallel & Distributed Processing Symposium, IEEE, May 2023, St. Petersburg, Florida, United States. ⟨hal-04013708⟩
72 View
81 Download

Share

Gmail Facebook X LinkedIn More