Overview of Distributed Linear Algebra on Hybrid Nodes over the StarPU Runtime - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Overview of Distributed Linear Algebra on Hybrid Nodes over the StarPU Runtime

Abstract

Runtime systems usually abstract a single node, like Plasma/Quark, Flame/SuperMatrix, Morse/StarPU,Dplasma/Parsec ... When going into harnessing cluster of nodes, how should they communicate? By using explicit MPI user calls ? By using a specific paradigm (Dplasma) ? Or can we keep the same paradigm and almost the same code, and leave the runtime system handle data transfers? We show how such a system has been sucessfully implemented on top of the StarPU runtime.
Fichier principal
Vignette du fichier
siampp14.pdf (299.15 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00978602 , version 1 (14-04-2014)

Identifiers

  • HAL Id : hal-00978602 , version 1

Cite

Emmanuel Agullo, Olivier Aumage, Mathieu Faverge, Nathalie Furmento, Florent Pruvost, et al.. Overview of Distributed Linear Algebra on Hybrid Nodes over the StarPU Runtime. SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP 2014), Feb 2014, Portland, Oregon, United States. ⟨hal-00978602⟩
263 View
202 Download

Share

Gmail Facebook Twitter LinkedIn More