PSBLAS: A library for parallel linear algebra computation on sparse matrices

Citation
S. Filippone et M. Colajanni, PSBLAS: A library for parallel linear algebra computation on sparse matrices, ACM T MATH, 26(4), 2000, pp. 527-550
Citations number
23
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Computer Science & Engineering
Journal title
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE
ISSN journal
0098-3500 → ACNP
Volume
26
Issue
4
Year of publication
2000
Pages
527 - 550
Database
ISI
SICI code
0098-3500(200012)26:4<527:PALFPL>2.0.ZU;2-P
Abstract
Many computationally intensive problems in engineering and science give ris e to the solution of large, sparse, linear systems of equations. Fast and e fficient methods for their solution are very important because these system s usually occur in the innermost loop of the computational scheme. Parallel ization is often necessary to achieve an acceptable level of performance. T his paper presents the design, implementation, and interface of a library o f Basic Linear Algebra Subroutines for sparse matrices (PSBLAS) which is sp ecifically tailored to distributed-memory computers. PSBLAS enables easy, e fficient, and portable implementations of parallel iterative solvers for li near systems. The interface keeps in view a Single Program Multiple Data pr ogramming model on distributed-memory machines. However, the architecture o f the library does not exclude an implementation in different paradigms, su ch as those based on the shared-memory model.