Evaluation of the Feedback Guided Dynamic Loop Scheduling (FGDLS) Algorithms

Sabin TABIRCA  Tatiana TABIRCA  Laurence T. YANG  Len FREEMAN  

IEICE TRANSACTIONS on Information and Systems   Vol.E87-D   No.7   pp.1829-1833
Publication Date: 2004/07/01
Online ISSN: 
Print ISSN: 0916-8532
Type of Manuscript: Special Section PAPER (Special Section on Hardware/Software Support for High Performance Scientific and Engineering Computing)
Category: Distributed, Grid and P2P Computing
parallel loop scheduling,  feedback-guided dynamic loop scheduling,  algorithms,  

Full Text: PDF(350.4KB)
>>Buy this Article

In this paper we consider the Feedback-Guided Dynamic Loop Scheduling (FGDLS) method that was proposed by Bull. The method uses a feedback-guided mechanism to schedule a parallel loop within a sequential outer loop. The execution times and the scheduling bounds at a outer iteration are used to find the scheduling bound of the next outer iteration. In this way FGDLS achieves an optimal load balance. Two algorithms have been proposed so far by Tabirca et al. In this article we will review these two algorithms and will give a comparison between their performances.