ePrints

The Performance of the BLAS and LAPACK on a Shared Memory Scalar Multiprocessor

Lookup NU author(s): Professor Chris Phillips

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

LAPACK is a set of Fortran subroutines covering a wide area of linear algebra algorithms. It was developed with the intention of being portable across a range of parallel processing environments. We consider here the performance of the Basic Linear Algebra Subprograms (BLAS) on which this software is based, on a shared memory multiprocessor, with each processing element possessing scalar characteristics only. We then investigate the effect of the block size on the performance of certain of the LAPACK routines when executed in the same environment.

Publication metadata

Author(s): Phillips C

Publication type: Article

Publication status: Published

Journal: Parallel Computing

Year: 1991

Volume: 17

Issue: 6-7

Pages: 751-761

ISSN (print): 0167-8191

ISSN (electronic): 1872-7336

Publisher: Elsevier BV

URL: http://dx.doi.org/10.1016/S0167-8191(05)80064-X

DOI: 10.1016/S0167-8191(05)80064-X

Altmetrics

Share