Performance evaluation of heuristics for scheduling pipelined multiprocessor tasks

M. Fikret Ercan, Ceyda Oguz, Yu Fai Fung

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

2 Citations (Scopus)

Abstract

This paper presents the evaluation of the solution quality of heuristic algorithms developed for scheduling multiprocessor tasks in a class of multiprocessor architecture designed for real-time operations. MIMD parallelism and multiprogramming support are the two main characteristics of multiprocessor architecture considered. The solution methodology includes different techniques including simulated annealing, tabu search, as well as well-known simple priority rule based heuristics. The results obtained by these different techniques are analyzed for different number of jobs and machine configurations.

Original languageEnglish
Title of host publicationComputational Science - ICCS 2001 - International Conference, 2001, Proceedings
EditorsVassil N. Alexandrov, Jack J. Dongarra, Benjoe A. Juliano, René S. Renner, C.J. Kenneth Tan
PublisherSpringer Verlag
Pages61-70
Number of pages10
ISBN (Print)3540422323, 9783540422327
Publication statusPublished - May 2001
EventInternational Conference on Computational Science, ICCS 2001 - San Francisco, United States
Duration: 28 May 200130 May 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2073
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceInternational Conference on Computational Science, ICCS 2001
Country/TerritoryUnited States
CitySan Francisco
Period28/05/0130/05/01

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Performance evaluation of heuristics for scheduling pipelined multiprocessor tasks'. Together they form a unique fingerprint.

Cite this