Efficient sparse LU factorization with left-right looking strategy on shared memory multiprocessors

被引:135
作者
Schenk, O
Gärtner, K
Fichtner, W
机构
[1] ETH Zurich, Swiss Fed Inst Technol, Integrated Syst Lab, CH-8092 Zurich, Switzerland
[2] Weierstr Inst Appl Anal & Stochast, DE-10117 Berlin, Germany
来源
BIT | 2000年 / 40卷 / 01期
关键词
parallel sparse LU factorization; asynchronous computation scheduling; SMP parallel computing; multigrid coarse solver; process simulation; device simulation;
D O I
10.1023/A:1022326604210
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
An efficient sparse LU factorization algorithm on popular shared memory multiprocessors is presented. Pipelining parallelism is essential to achieve higher parallel efficiency and it is exploited with a left-right looking algorithm. No global barrier is used and a completely asynchronous scheduling scheme is one central point of the implementation. The algorithm has been successfully tested on SUN Enterprise, DEC AlphaServer, SGI Origin 2000 and Cray T90 and J90 parallel computers, delivering up to 2.3 GFlop/s on an eight processor DEC AlphaServer for medium-size semiconductor device simulations and structural engineering problems.
引用
收藏
页码:158 / 176
页数:19
相关论文
共 35 条
[1]  
AMESTOY P, IN PRESS COMPUT METH
[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]  
ASHCRAFT CC, 1987, INT J SUPERCOMPUT AP, V1, P10
[4]  
BARNARD S, 1994, RNR92033 NASA AM RES
[5]  
BUNCH JR, 1977, MATH COMPUT, V31, P162
[6]  
DAGNUM RML, 1998, IEEE COMPUT SCI ENG, V1, P46
[7]   LEVEL-3 BLAS IN LU FACTORIZATION ON THE CRAY-2, ETA-10P, AND IBM 3090-200 VF [J].
DAYDE, MJ ;
DUFF, IS .
INTERNATIONAL JOURNAL OF SUPERCOMPUTER APPLICATIONS AND HIGH PERFORMANCE COMPUTING, 1989, 3 (02) :40-70
[8]  
DEMMEL J, 1997, CSD97943 U CAL COMP
[9]  
DONGARRA J, 1991, SUPERCOMPUTER, V8, P33
[10]  
DONGARRA J, 1998, CS8985 U TENN DEP CO