PERFORMANCE OF LAPACK - A PORTABLE LIBRARY OF NUMERICAL LINEAR ALGEBRA ROUTINES

被引：11

作者：

ANDERSON, EC ^{[1
]}

DONGARRA, J ^{[1
]}

机构：

[1] UNIV TENNESSEE,DEPT COMP SCI,KNOXVILLE,TN 37996

来源：

PROCEEDINGS OF THE IEEE | 1993年 / 81卷 / 08期

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/5.236187

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper describes the LAPACK project, an effort to produce a numerical linear algebra library that runs efficiently on shared memory vector and parallel processors. A description is given on what was done to achieve performance and results are given for various computers. In addition, future directions for research on parallel computers are discussed.

引用

页码：1094 / 1102

页数：9

共 16 条

[1]

Anderson E., 1990, P 1990 ACM IEEE C SU, P2

[2]

Anderson E., 1992, LAPACK USERS GUIDE

[3]

ANDERSON E, 1989, CS8989 U TENN TECH R

[4]

ANDERSON E, 1990, 4TH P SIAM C PAR PRO, P3

[5]

Bischof C. H., 1989, Journal of Supercomputing, V3, P193, DOI 10.1007/BF00127828

[6]

BISCHOF CH, 1990, LECT NOTES COMPUT SC, V457, P210

[7]

Dongarra J. J., 1979, LINPACK USERS GUIDE

[8] AN EXTENDED SET OF FORTRAN BASIC LINEAR ALGEBRA SUBPROGRAMS [J].

DONGARRA, JJ ;

DUCROZ, J ;

HAMMARLING, S ;

HANSON, RJ .

ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 1988, 14 (01) :1-17

[9] DISTRIBUTION OF MATHEMATICAL SOFTWARE VIA ELECTRONIC MAIL [J].

DONGARRA, JJ ;

GROSSE, E .

COMMUNICATIONS OF THE ACM, 1987, 30 (05) :403-407

[10]

DONGARRA JJ, 1990, ACM T MATH SOFTWARE, V16, P1, DOI 10.1145/77626.79170

← 1 2 →