Dense Linear Algebra Kernels on Heterogeneous Platforms: Redistribution Issues

Olivier Beaumont 1 Arnaud Legrand 1 Fabrice Rastello 1 Yves Robert 1
1 REMAP - Regularity and massive parallel computing
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : Redistribution algorithms for dense linear algebra kernels on heterogeneous platforms are considered. In this context, processor speeds may well vary during the execution of a large kernel, which requires efficient strategies for redistributing the data along the computations. The proposed strategy is to redistribute data after some well-identified static phases and therefore is neither fully static nor fully dynamic. An optimal algorithm (under some assumptions) for redistributing data when computing the product of two matrices is presented.
Complete list of metadatas

https://hal.inria.fr/hal-00789434
Contributor : Arnaud Legrand <>
Submitted on : Monday, February 18, 2013 - 11:51:05 AM
Last modification on : Friday, April 20, 2018 - 3:44:24 PM

Identifiers

Collections

Citation

Olivier Beaumont, Arnaud Legrand, Fabrice Rastello, Yves Robert. Dense Linear Algebra Kernels on Heterogeneous Platforms: Redistribution Issues. Parallel Computing, Elsevier, 2002, 28, pp.155―185. ⟨10.1016/S0167-8191(01)00134-X⟩. ⟨hal-00789434⟩

Share

Metrics

Record views

224