Towards an implementation of a multilevel ILU preconditioner on shared-memory computers

Arnold Meijster; Fred Wubs

Towards an implementation of a multilevel ILU preconditioner on shared-memory computers

Arnold Meijster, Fred Wubs

Computational and Numerical Mathematics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)

236 Downloads (Pure)

Abstract

Recently, substantial progress has been made in the development of multilevel ILU-factorizations. These methods are attractive for very large problems due to their good convergence properties. We consider the parallelization of the instance MRILU, where we restrict to a version intended for scalar problems. The most time consuming parts in using MRILU are repeated multiplication of two sparse matrices in the construction phase and the multiplication of a sparse matrix and a full vector in the solution phase. Algorithms for these operations, as well as matrix transposition, are presented and have been tested on a Cray J90.

Original language	English
Title of host publication	HIGH PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS
Editors	M Bubak, R Williams, H Afsarmanesh, B Hertzberger
Place of Publication	BERLIN
Publisher	Springer
Pages	109-118
Number of pages	10
ISBN (Print)	3-540-67553-1
Publication status	Published - 2000
Event	8th International Conference on High Performance Computing and Networking - , Netherlands Duration: 8-May-2000 → 10-May-2000

Publication series

Name	LECTURE NOTES IN COMPUTER SCIENCE
Publisher	SPRINGER-VERLAG BERLIN
Volume	1823
ISSN (Print)	0302-9743

Other

Other	8th International Conference on High Performance Computing and Networking
Country	Netherlands
Period	08/05/2000 → 10/05/2000

Keywords

SPARSE MATRICES

Access to Document

2000LNCSMeijster.pdfFinal publisher's version, 155 KB

Cite this

@inproceedings{a0290b66148a4074a0101e9ba7dcbdb6,

title = "Towards an implementation of a multilevel ILU preconditioner on shared-memory computers",

abstract = "Recently, substantial progress has been made in the development of multilevel ILU-factorizations. These methods are attractive for very large problems due to their good convergence properties. We consider the parallelization of the instance MRILU, where we restrict to a version intended for scalar problems. The most time consuming parts in using MRILU are repeated multiplication of two sparse matrices in the construction phase and the multiplication of a sparse matrix and a full vector in the solution phase. Algorithms for these operations, as well as matrix transposition, are presented and have been tested on a Cray J90.",

keywords = "SPARSE MATRICES",

author = "Arnold Meijster and Fred Wubs",

note = "Relation: http://www.rug.nl/informatica/organisatie/overorganisatie/iwi Rights: University of Groningen, Research Institute for Mathematics and Computing Science (IWI); 8th International Conference on High Performance Computing and Networking ; Conference date: 08-05-2000 Through 10-05-2000",

year = "2000",

language = "English",

isbn = "3-540-67553-1",

series = "LECTURE NOTES IN COMPUTER SCIENCE",

publisher = "Springer",

pages = "109--118",

editor = "M Bubak and R Williams and H Afsarmanesh and B Hertzberger",

booktitle = "HIGH PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS",

}

Meijster, A & Wubs, F 2000, Towards an implementation of a multilevel ILU preconditioner on shared-memory computers. in M Bubak, R Williams, H Afsarmanesh & B Hertzberger (eds), HIGH PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS. LECTURE NOTES IN COMPUTER SCIENCE, vol. 1823, Springer, BERLIN, pp. 109-118, 8th International Conference on High Performance Computing and Networking, Netherlands, 08/05/2000.

Towards an implementation of a multilevel ILU preconditioner on shared-memory computers. / Meijster, Arnold ; Wubs, Fred.

HIGH PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS. ed. / M Bubak; R Williams; H Afsarmanesh; B Hertzberger. BERLIN : Springer, 2000. p. 109-118 (LECTURE NOTES IN COMPUTER SCIENCE; Vol. 1823).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Towards an implementation of a multilevel ILU preconditioner on shared-memory computers

AU - Meijster, Arnold

AU - Wubs, Fred

N1 - Relation: http://www.rug.nl/informatica/organisatie/overorganisatie/iwi Rights: University of Groningen, Research Institute for Mathematics and Computing Science (IWI)

PY - 2000

Y1 - 2000

N2 - Recently, substantial progress has been made in the development of multilevel ILU-factorizations. These methods are attractive for very large problems due to their good convergence properties. We consider the parallelization of the instance MRILU, where we restrict to a version intended for scalar problems. The most time consuming parts in using MRILU are repeated multiplication of two sparse matrices in the construction phase and the multiplication of a sparse matrix and a full vector in the solution phase. Algorithms for these operations, as well as matrix transposition, are presented and have been tested on a Cray J90.

AB - Recently, substantial progress has been made in the development of multilevel ILU-factorizations. These methods are attractive for very large problems due to their good convergence properties. We consider the parallelization of the instance MRILU, where we restrict to a version intended for scalar problems. The most time consuming parts in using MRILU are repeated multiplication of two sparse matrices in the construction phase and the multiplication of a sparse matrix and a full vector in the solution phase. Algorithms for these operations, as well as matrix transposition, are presented and have been tested on a Cray J90.

KW - SPARSE MATRICES

M3 - Conference contribution

SN - 3-540-67553-1

T3 - LECTURE NOTES IN COMPUTER SCIENCE

SP - 109

EP - 118

BT - HIGH PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS

A2 - Bubak, M

A2 - Williams, R

A2 - Afsarmanesh, H

A2 - Hertzberger, B

PB - Springer

CY - BERLIN

T2 - 8th International Conference on High Performance Computing and Networking

Y2 - 8 May 2000 through 10 May 2000

ER -