Treffer: Analysis and comparison of two general sparse solvers for distributed memory computers

Title:
Analysis and comparison of two general sparse solvers for distributed memory computers
Authors:
Amestoy, Patrick, Duff, Iain S., L'Excellent, Jean-Yves, Li, Xiaoye S.
Contributors:
Algorithmes Parallèles et Optimisation (IRIT-APO), Institut de recherche en informatique de Toulouse (IRIT), Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Toulouse Mind & Brain Institut (TMBI), Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT), Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique (CERFACS), Lawrence Berkeley National Laboratory Berkeley (LBNL)
Source:
ACM Transactions on Mathematical Software. 27:388-421
Publisher Information:
Association for Computing Machinery (ACM), 2001.
Publication Year:
2001
Subject Terms:
[INFO.INFO-DC]Computer Science [cs]/Distributed, [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS], Parallel numerical computation, 02 engineering and technology, preorderings, Direct numerical methods for linear systems and matrix inversion, MUMPS, Parallel, large sparse sets of linear equations, 01 natural sciences, Computational methods for sparse matrices, and Cluster Computing [cs.DC], large-scale distributed-memory computers, numerical stability, 0202 electrical engineering, electronic engineering, information engineering, Analysis of algorithms, 0101 mathematics, performance
Document Type:
Fachzeitschrift Article
File Description:
application/xml
Language:
English
ISSN:
1557-7295
0098-3500
DOI:
10.1145/504210.504212
Access URL:
http://epubs.cclrc.ac.uk/bitstream/259/raltr-2001003.pdf
Rights:
URL: https://www.acm.org/publications/policies/copyright_policy#Background
Accession Number:
edsair.doi.dedup.....9c0e04027c9da807b88c75c84c6d8b14
Database:
OpenAIRE

Weitere Informationen

This paper provides a comprehensive study and comparison of two state-of-the-art direct solvers for large sparse sets of linear equations on large-scale distributed-memory computers. One is a multifrontal solver called MUMPS, the other is a supernodal solver called superLU. We describe the main algorithmic features of the two solvers and compare their performance characteristics with respect to uniprocessor speed, interprocessor communication, and memory requirements. For both solvers, preorderings for numerical stability and sparsity play an important role in achieving high parallel efficiency. We analyse the results with various ordering algorithms. Our performance analysis is based on data obtained from runs on a 512-processor Cray T3E using a set of matrices from real applications. We also use regular 3D grid problems to study the scalability of the two solvers.