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

被引：248

作者：

Agullo, Emmanuel ^{[1
]}

Demmel, Jim ^{[1
]}

Dongarra, Jack ^{[1
]}

Hadri, Bilel ^{[1
]}

Kurzak, Jakub ^{[1
]}

Langou, Julien ^{[1
]}

Ltaief, Hatem ^{[1
]}

Luszczek, Piotr ^{[1
]}

Tomov, Stanimire ^{[1
]}

机构：

[1] Univ Tennessee, Dept Elect Engn & Comp Sci, Knoxville, TN 37996 USA

来源：

SCIDAC 2009: SCIENTIFIC DISCOVERY THROUGH ADVANCED COMPUTING | 2009年 / 180卷

关键词：

D O I：

10.1088/1742-6596/180/1/012037

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

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.

引用

页数：5

共 12 条

[1]

AGULLO E, 2009, SC09 UNPUB

[2]

Anderson E., 1992, LAPACK Users Guide

[3]

[Anonymous], BLAS BASIC LINEAR AL

[4]

[Anonymous], 1986, Graph Theory: An Algorithmic Approach

[5]

[Anonymous], 2006, Tech. rep.

[6]

BABOULIN M, 2008, SUP 08 NOV 18

[7]

Blackford L., 1997, ScaLAPACK Users Guide

[8]

Buttari A., 2006, PARA 2006 UM SWED JU

[9] A class of parallel tiled linear algebra algorithms for multicore architectures [J].

Buttari, Alfredo ;

Langou, Julien ;

Kurzak, Jakub ;

Dongarra, Jack .

PARALLEL COMPUTING, 2009, 35 (01) :38-53

[10]

Song F., 2009, 638 UTK CS

← 1 2 →