Enhancing self-scheduling algorithms via synchronization and weighting

Ciorba, FM; Riakiotakis, I; Andronikos, T; Papakonstantinou, G; Chronopoulos, AT

dc.contributor.author	Ciorba, FM	en
dc.contributor.author	Riakiotakis, I	en
dc.contributor.author	Andronikos, T	en
dc.contributor.author	Papakonstantinou, G	en
dc.contributor.author	Chronopoulos, AT	en
dc.date.accessioned	2014-03-01T01:28:16Z
dc.date.available	2014-03-01T01:28:16Z
dc.date.issued	2008	en
dc.identifier.issn	0743-7315	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/18790
dc.subject	Dynamic load balancing algorithms	en
dc.subject	Loops with dependencies	en
dc.subject	Non-dedicated heterogeneous systems	en
dc.subject	Synchronization	en
dc.subject	Weighting	en
dc.subject.classification	Computer Science, Theory & Methods	en
dc.subject.other	Program processors	en
dc.subject.other	Resource allocation	en
dc.subject.other	Scheduling	en
dc.subject.other	Synchronization	en
dc.subject.other	Dynamic load balancing algorithms	en
dc.subject.other	Inter processor communication	en
dc.subject.other	Non dedicated heterogeneous systems	en
dc.subject.other	Scheduling algorithms	en
dc.title	Enhancing self-scheduling algorithms via synchronization and weighting	en
heal.type	journalArticle	en
heal.identifier.primary	10.1016/j.jpdc.2007.07.003	en
heal.identifier.secondary	http://dx.doi.org/10.1016/j.jpdc.2007.07.003	en
heal.language	English	en
heal.publicationDate	2008	en
heal.abstract	Existing dynamic self-scheduling algorithms, used to schedule independent tasks on heterogeneous clusters, cannot handle tasks with dependencies because they lack the support for internode communication. To compensate for this deficiency we introduce a synchronization mechanism that provides inter-processor communication, thus, enabling self-scheduling algorithms to handle efficiently nested loops with dependencies. We also present a weighting mechanism that significantly improves the performance of dynamic self-scheduling algorithms. These algorithms divide the total number of tasks into chunks and assign them to processors. The weighting mechanism adapts the chunk sizes to the computing power and current run-queue state of the processors. The synchronization and weighting mechanisms are orthogonal, in the sense that they can simultaneously be applied to loops with dependencies. Thus, they broaden the application spectrum of dynamic self-scheduling algorithms and improve their performance. Extensive testing confirms the efficiency of the synchronization and weighting mechanisms and the significant improvement of the synchronized-weighted versions of the algorithms over the synchronized-only versions. (c) 2007 Elsevier Inc. All rights reserved.	en
heal.publisher	ACADEMIC PRESS INC ELSEVIER SCIENCE	en
heal.journalName	Journal of Parallel and Distributed Computing	en
dc.identifier.doi	10.1016/j.jpdc.2007.07.003	en
dc.identifier.isi	ISI:000253282300010	en
dc.identifier.volume	68	en
dc.identifier.issue	2	en
dc.identifier.spage	246	en
dc.identifier.epage	264	en