Multi-Objective Scheduling for Parallel Jobs on Grid

Abstract:

Article Preview

Focused on the complexity of the parallel job scheduling on heterogeneous Grid, the paper proposes a multi-objective optimization based scheduling algorithm. The algorithm first splits the parallel job up into a series of independent processes with constraints, and then adopts particles to represent the mapping of job-resource. Multi-objective PSO is employed to simultaneously optimize the scheduling objectives of throughput and average turnaround time. Experimental result indicates that the proposed approach is effective while dealing with large scale parallel jobs scheduling on heterogeneous Grid and outperforms other conventional algorithms.

Info:

Periodical:

Key Engineering Materials (Volumes 439-440)

Edited by:

Yanwen Wu

Pages:

1281-1286

DOI:

10.4028/www.scientific.net/KEM.439-440.1281

Citation:

P. F. Liu and S. B. Dong, "Multi-Objective Scheduling for Parallel Jobs on Grid", Key Engineering Materials, Vols. 439-440, pp. 1281-1286, 2010

Online since:

June 2010

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.