Browse by author
Lookup NU author(s): Professor Chris Phillips
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
We describe an attempt to improve the performance of a sequential Fortran code on a shared memory multiprocessor. The only parallelism employed is that provided by a set of parallel BLAS. The code is modified to make maximum use of the BLAS and, in particular, to use an appropriate LAPACK routine. Comparisons are made with automatically generated parallel code.
Author(s): Beavis AW, Phillips C
Publication type: Article
Publication status: Published
Journal: Concurrency: Practice and Experience
Year: 1992
Volume: 4
Issue: 8
Pages: 575-587
ISSN (print): 1040-3108
ISSN (electronic): 1532-0634
Publisher: John Wiley & Sons Ltd.
URL: http://dx.doi.org/10.1002/cpe.4330040802
DOI: 10.1002/cpe.4330040802
Altmetrics provided by Altmetric