dc.contributor.author |
GEORGAKOPOULOS, G |
en |
dc.contributor.author |
STAFYLOPATIS, A |
en |
dc.date.accessioned |
2014-03-01T01:09:17Z |
|
dc.date.available |
2014-03-01T01:09:17Z |
|
dc.date.issued |
1993 |
en |
dc.identifier.issn |
0988-3754 |
en |
dc.identifier.uri |
https://dspace.lib.ntua.gr/xmlui/handle/123456789/10876 |
|
dc.relation.uri |
http://www.informatik.uni-trier.de/~ley/db/journals/ita/ita27.html#GeorgakopoulosS93 |
en |
dc.subject |
Parallel Algorithm |
en |
dc.subject.classification |
Computer Science, Information Systems |
en |
dc.title |
AN APPROACH TO PARALLEL ALGORITHM DESIGN |
en |
heal.type |
journalArticle |
en |
heal.language |
English |
en |
heal.publicationDate |
1993 |
en |
heal.abstract |
Effort dedicated to the analysis of parallel computation has mainly focused in the direction of finding parallel algorithms for specific problems. On the other hand, there is a need for general purpose tools for ''automatically'' assigning complex tasks to parallel computing machines (compilers for parallel machines provide an example). In this paper, we describe an approach for finding an optimal, in some sense, schedule for assigning tasks to a set of parallel processors. Finding this schedule remains an NP-complete problem in the general case, but our definition allows us to solve it in polynomial time for a large class of task graphs and a large class of machine architectures. Point symmetric graphs and their generalization play an important role, and some of their properties are discussed. |
en |
heal.publisher |
DUNOD |
en |
heal.journalName |
RAIRO-INFORMATIQUE THEORIQUE ET APPLICATIONS-THEORETICAL INFORMATICS AND APPLICATIONS |
en |
dc.identifier.isi |
ISI:A1993KW95300001 |
en |
dc.identifier.volume |
27 |
en |
dc.identifier.issue |
2 |
en |
dc.identifier.spage |
85 |
en |
dc.identifier.epage |
95 |
en |