De-centralized Job Scheduling on Computational Grids Using Distributed Backfilling

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

3 Scopus citations

Abstract

To dynamically improve node selections of the waiting jobs under de-centralized scheduling, each node together with its neighbors is assumed to compose a subgrid, using distributed backfilling to optimize grid scheduling on each subgrid. Whenever a job terminates, distributed backfilling is triggered to rebackfill all waiting jobs on the corresponding subgrid in order of their submittal. Each subgrid is overlapped with some other, so the waiting jobs may be migrated around the grid. A simulated grid is established, while grid workload is modeled by extending workload models of parallel systems. Job speedup is used to evaluate scheduling strategies. Results show the dynamic optimization of node selections brought by distributed backfilling is grid-wide, and can improve scheduling performance remarkably as long as grid load is not too light and the job migration costs are not too high.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsHai Jin, Jianhua Sun, Yi Pan, Nong Xiao
PublisherSpringer Verlag
Pages285-292
Number of pages8
ISBN (Print)3540235647, 9783540235644
DOIs
StatePublished - 2004

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3251
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Dive into the research topics of 'De-centralized Job Scheduling on Computational Grids Using Distributed Backfilling'. Together they form a unique fingerprint.

Cite this