DocumentCode
3042912
Title
Memory-based scheduling for a parallel multifrontal solver
Author
Guermouche, Abdou ; Excellent, Jean-Yves L.
Author_Institution
INRIA GRAAL Project, Ecole Normale Superieure de Lyon, France
fYear
2004
fDate
26-30 April 2004
Firstpage
71
Abstract
Summary form only given. The memory usage of sparse direct solvers can be the bottleneck to solve large-scale problems. We describe dynamic scheduling strategies that aim at reducing the memory usage of a parallel direct solver. Combined to static modifications of the tasks dependency graph, experiments show that such techniques have a good potential to improve the memory usage of a parallel multifrontal solver, MUMPS.
Keywords
dynamic scheduling; parallel memories; MUMPS parallel multifrontal solver; memory-based scheduling; parallel direct solver; sparse direct solvers; tasks dependency graph; Assembly; Dynamic scheduling; Equations; Iterative methods; Large-scale systems; Memory management; Processor scheduling; Robustness; Scalability; Sparse matrices;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel and Distributed Processing Symposium, 2004. Proceedings. 18th International
Print_ISBN
0-7695-2132-0
Type
conf
DOI
10.1109/IPDPS.2004.1303001
Filename
1303001
Link To Document