A parallel block LU decomposition method for distributed finite element matrices

被引：14
作者：
Maurer, Daniel [1 ]
Wieners, Christian [1 ]
机构：
[1] Karlsruhe Inst Technol, Inst Appl & Numer Math 3, D-76128 Karlsruhe, Germany
来源：
PARALLEL COMPUTING | 2011年 / 37卷 / 12期
关键词：
Parallel computing; Finite elements; Direct solver for linear equations; Block LU decomposition; STABILITY; SOLVER; ALGORITHMS; SIMULATION; PARDISO; SYSTEMS;
D O I：
10.1016/j.parco.2011.05.007
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
In this work we present a new parallel direct linear solver for matrices resulting from finite element problems. The algorithm follows the nested dissection approach, where the resulting Schur complements are also distributed in parallel. The sparsity structure of the finite element matrices is used to pre-compute an efficient block structure for the LU factors. We demonstrate the performance and the parallel scaling behavior by several test examples. (C) 2011 Elsevier B.V. All rights reserved.
引用
收藏
页码：742 / 758
页数：17
相关论文
共 32 条
[1]   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
[2]  
Anderson E., 1992, LAPACKS USERS GUIDE
[3]  
[Anonymous], P 1995 ACM IEEE C SU
[4]  
[Anonymous], 1990, LAPACK PORTABLE LINE
[5]  
[Anonymous], MUMPS: a multifrontal massively parallel sparse direct solver
[6]  
[Anonymous], 1997, THEORY FAST SOLVERS
[7]  
Bomhof CW, 2000, NUMER LINEAR ALGEBR, V7, P649, DOI 10.1002/1099-1506(200010/12)7:7/8<649::AID-NLA217>3.0.CO
[8]  
2-W
[9]  
*CST, CST COMP SIM TECHN D
[10]  
DACKLAND K, 1992, PROCEEDINGS OF THE FIFTH SIAM CONFERENCE ON PARALLEL PROCESSING FOR SCIENTIFIC COMPUTING, P3
1234