An Implementation of the Matrix Multiplication Algorithm SUMMA in mpF

In this paper, we present a new parallel Fortran extension called mpF. The language based on both data and task parallelism allows explicit specification of data and computations distribution. We discuss some reasons for the language design and demonstrate the basic mpF features on an example of the...

Full description

Saved in:
Bibliographic Details
Published inParallel Computing Technologies pp. 420 - 432
Main Authors Kalinov, Alexey, Ledovskikh, Ilya, Posypkin, Mikhail, Levchenko, Zakhar, Chizhov, Vladimir
Format Book Chapter
LanguageEnglish
Published Berlin, Heidelberg Springer Berlin Heidelberg 2005
SeriesLecture Notes in Computer Science
Subjects
Computing Space
High Performance Fortran
Matrix Multiplication Algorithm
Message Passing Interface
Task Parallelism
Online AccessGet full text
ISBN3540281266
9783540281269
ISSN0302-9743
1611-3349
DOI10.1007/11535294_37

Cover

More Information
Summary:In this paper, we present a new parallel Fortran extension called mpF. The language based on both data and task parallelism allows explicit specification of data and computations distribution. We discuss some reasons for the language design and demonstrate the basic mpF features on an example of the parallel matrix multiplication algorithm SUMMA. The mpF implementation is compared with its MPI counterpart.
Bibliography:This research is supported by Computational and Information Aspects of Solving Large Problems program of the Division of Mathematical Sciences of the Russian Academy of Sciences.
ISBN:3540281266
9783540281269
ISSN:0302-9743
1611-3349
DOI:10.1007/11535294_37