Data-flow algorithms for parallel matrix computation

In this article we develop some algorithms and tools for solving matrix problems on parallel processing computers. Operations are synchronized through data-flow alone, which makes global synchronization unnecessary and enables the algorithms to be implemented on machines with very simple operating s...

Full description

Saved in:
Bibliographic Details
Published inCommunications of the ACM Vol. 28; no. 8; pp. 840 - 853
Main Authors O'Leary, Dianne P., Stewart, G. W.
Format Magazine Article
LanguageEnglish
Published 01.08.1985
Online AccessGet full text
ISSN0001-0782
1557-7317
1557-7317
DOI10.1145/4021.4025

Cover

More Information
Summary:In this article we develop some algorithms and tools for solving matrix problems on parallel processing computers. Operations are synchronized through data-flow alone, which makes global synchronization unnecessary and enables the algorithms to be implemented on machines with very simple operating systems and communication protocols. As examples, we present algorithms that form the main modules for solving Liapounov matrix equations. We compare this approach to wave front array processors and systolic arrays, and note its advantages in handling missized problems, in evaluating variations of algorithms or architectures, in moving algorithms from system to system, and in debugging parallel algorithms on sequential machines.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0001-0782
1557-7317
1557-7317
DOI:10.1145/4021.4025