An efficient evolutionary scheduling algorithm for parallel job model in grid environment

Authors:
Piotr Switalski;Franciszek Seredynski
Affiliations:
Siedlce University of Natural Sciences and Humanities, Computer Science Department, Siedlce, Poland;Polish-Japanese Institute of Information Technology, Warsaw, Poland and Institute of Computer Science, Polish Academy of Sciences, Warsaw, Poland
Venue:
PaCT'11 Proceedings of the 11th international conference on Parallel computing technologies
Year:
2011

Citing 6
Cited 1

A Distributed Heterogeneous Supercomputing Management System

Computer
On Advantages of Grid Computing for Parallel Job Scheduling

CCGRID '02 Proceedings of the 2nd IEEE/ACM International Symposium on Cluster Computing and the Grid
Applying economic scheduling methods to Grid environments

Grid resource management
The Anatomy of the Grid: Enabling Scalable Virtual Organizations

International Journal of High Performance Computing Applications
A comparison between two grid scheduling philosophies: EGEE WMS and Grid Way

Multiagent and Grid Systems - Grid Computing, high performance and distributed applications
On-line hierarchical job scheduling on grids with admissible allocation

Journal of Scheduling

Parallel evaluation of Pittsburgh rule-based classifiers on GPUs

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we propose an efficient parallel job scheduling algorithm for a grid environment. The model implies two stage scheduling. At the first stage, algorithm allocates jobs to the suitable machines, where at the second stage jobs are independently scheduled on each machine. Allocation of jobs on the first stage of the algorithm is performed with use of a relatively new evolutionary algorithm called Generalized Extremal Optimization (GEO). GEO is inspired by a simple coevolutionary model known as Bak-Sneppen model. Scheduling on the second stage is performed by some proposed heuristic. We compare GEO-based scheduling algorithm applied on the first stage with Genetic Algorithm (GA)-based scheduling algorithm. Experimental results show that the GEO, despite of its simplicity, outperforms the GA algorithm in all range of scheduling instances.