Matrices over runtime systems at exascale

Emmanuel Agullo, George Bosilca, Bérenger Bramas, Cedric Castagnede, Olivier Coulaud, Eric F. Darve, Jack Dongarra, Mathieu Faverge, Nathalie Furmento, Luc Giraud, Xavier Lacoste, Julien Langou, Hatem Ltaief, Matthias Messner, Raymond Namyst, Pierre Ramet, Toru Takahashi, Samuel Thibault, Stanimire Z. Tomov, Ichitaro Yamazaki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

The goal of Matrices Over Runtime Systems at Exascale (MORSE) project is to design dense and sparse linear algebra methods that achieve the fastest possible time to an accurate solution on large-scale multicore systems with GPU accelerators, using all the processing power that future high end systems can make available. In this poster, we propose a framework for describing linear algebra algorithms at a high level of abstraction and delegating the actual execution to a runtime system in order to design software whose performance is portable accross architectures. We illustrate our methodology on three classes of problems: dense linear algebra, sparse direct methods and fast multipole methods. The resulting codes have been incorporated into Magma, Pastix and ScalFMM solvers, respectively. © 2012 IEEE.
Original languageEnglish (US)
Title of host publication2012 SC Companion: High Performance Computing, Networking Storage and Analysis
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1330-1332
Number of pages3
ISBN (Print)9780769549569
DOIs
StatePublished - Nov 2012

Fingerprint Dive into the research topics of 'Matrices over runtime systems at exascale'. Together they form a unique fingerprint.

Cite this