article Open AccessTop 1% cited

Numerical linear algebra on emerging architectures: The PLASMA and MAGMA projects

Journal of Physics Conference Series · 2009 · Vol. 180 · pp. 012037–012037

Emmanuel Agullo✉(University of Tennessee at Knoxville)Jim Demmel(University of Tennessee at Knoxville)Jack Dongarra(University of Tennessee at Knoxville)Bilel Hadri(University of Tennessee at Knoxville)Jakub Kurzak(University of Tennessee at Knoxville)Julien Langou(University of Tennessee at Knoxville)Hatem Ltaief(University of Tennessee at Knoxville)Piotr Łuszczek(University of Tennessee at Knoxville)Stanimire Tomov(U.S. National Science Foundation)

View via DOI Publisher page

Abstract

The emergence and continuing use of multi-core architectures and graphics processing units require changes in the existing software and sometimes even a redesign of the established algorithms in order to take advantage of now prevailing parallelism. Parallel Linear Algebra for Scalable Multi-core Architectures (PLASMA) and Matrix Algebra on GPU and Multics Architectures (MAGMA) are two projects that aims to achieve high performance and portability across a wide range of multi-core architectures and hybrid systems respectively. We present in this document a comparative study of PLASMA's performance against established linear algebra packages and some preliminary results of MAGMA on hybrid multi-core and GPU systems.

Parallel Computing and Optimization Techniques Matrix Theory and Algorithms Interconnection Networks and SystemsSoftware portabilityLinear algebraComputer scienceMagmaGraphicsScalabilityNumerical linear algebraParallel computingSoftwareParallelism (grammar)

Citations

421

FWCI

21.69

field-weighted impact

References

Percentile

100%

vs. same field & year

Citations per year

References

Graph Theory: An Algorithmic Approach

Journal of the Operational Research Society · 1976 · 910 citations

Citation Network

How this paper connects to the literature. Drag to explore, click any node to open that paper.