Recent advances in direct methods for solving unsymmetric sparse systems of linear equations

被引:53
作者
Gupta, A [1 ]
机构
[1] IBM Corp, Thomas J Watson Res Ctr, Yorktown Hts, NY 10598 USA
来源
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE | 2002年 / 28卷 / 03期
关键词
algorithms; performance; sparse matrix factorization; sparse LU decomposition; multifrontal method; parallel sparse solvers;
D O I
10.1145/569147.569149
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
During the past few years, algorithmic improvements alone have reduced the time required for the direct solution of unsymmetric sparse systems of linear equations by almost an order of magnitude. This paper compares the performance of some well-known software packages for solving general sparse systems. In particular, it demonstrates the consistently high level of performance achieved by WSMP-the most recent of such solvers. It compares the various algorithmic components of these solvers and discusses their impact on solver performance. Our experiments show that the algorithmic choices made in WSMP enable it to run more than twice as fast as the best among similar solvers and that WSMP can factor some of the largest sparse matrices available from real applications in only a few seconds on a 4-CPU workstation. Thus, the combination of advances in hardware and algorithms makes it possible to solve those general sparse linear systems quickly and easily that might have been considered too large until recently.
引用
收藏
页码:301 / 324
页数:24
相关论文
共 45 条
[1]   An approximate minimum degree ordering algorithm [J].
Amestoy, PR ;
Davis, TA ;
Duff, IS .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 1996, 17 (04) :886-905
[2]   MEMORY MANAGEMENT ISSUES IN SPARSE MULTIFRONTAL METHODS ON MULTIPROCESSORS [J].
AMESTOY, PR ;
DUFF, IS .
INTERNATIONAL JOURNAL OF SUPERCOMPUTER APPLICATIONS AND HIGH PERFORMANCE COMPUTING, 1993, 7 (01) :64-82
[3]   Multifrontal parallel distributed symmetric and unsymmetric solvers [J].
Amestoy, PR ;
Duff, IS ;
L'Excellent, JY .
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2000, 184 (2-4) :501-520
[4]   VECTORIZATION OF A MULTIPROCESSOR MULTIFRONTAL CODE [J].
AMESTOY, PR ;
DUFF, IS .
INTERNATIONAL JOURNAL OF SUPERCOMPUTER APPLICATIONS AND HIGH PERFORMANCE COMPUTING, 1989, 3 (03) :41-59
[5]   A fully asynchronous multifrontal solver using distributed dynamic scheduling [J].
Amestoy, PR ;
Duff, IS ;
L'Excellent, JY ;
Koster, J .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2001, 23 (01) :15-41
[6]  
AMESTOY PR, 2001, ACM T MATH SOFTWARE, V27, P1
[7]  
AMESTOY PR, 2000, RTAPO003 ENSEEIHT IR
[8]  
ASHCRAFT C, 1996, 9601 CS YORK U DEP C
[9]  
Ashcraft C., 1999, PPSC
[10]  
COSNARD M, 2000, P INT PAR DISTR PROC