Title: Modified genetic algorithm for multiobjective task scheduling on heterogeneous computing system

Authors: O.L. Sathappan, P. Chitra, P. Venkatesh, M. Prabhu

Addresses: Department of Computer Science and Engineering, Thiagarajar College of Engineering, Madurai – 625015, Tamilnadu, India. ' Department of Computer Science and Engineering, Thiagarajar College of Engineering, Madurai – 625015, Tamilnadu, India. ' Department of Electrical and Electronics Engineering, Thiagarajar College of Engineering, Madurai – 625015, Tamilnadu, India. ' Department of Computer Science and Engineering, Thiagarajar College of Engineering, Madurai – 625015, Tamilnadu, India

Abstract: This paper addresses the problem of task scheduling in heterogeneous distributed systems with the goal of maximising the system reliability and decreasing the makespan. The task scheduling problem in heterogeneous systems an NP-complete problem. A modified genetic algorithm which combines the characteristics of bacteriological algorithm (BA) and genetic algorithm is proposed. This modified algorithm (MA) is used in the weighted sum approach of multiobjective genetic algorithm (MOGA). The proposed algorithm is applied for random and real-time numerical application graphs and compared with the biobjective genetic algorithm (BGA) in the literature. The simulation results confirm that the proposed algorithm produces near optimal solutions at reduced computational times.

Keywords: task scheduling; makespan; reliability; bacteriological algorithms; multiobjective genetic algorithms; MOGA; biobjective genetic algorithms; BGA; heterogeneous distributed systems.

DOI: 10.1504/IJITCC.2011.039282

International Journal of Information Technology, Communications and Convergence, 2011 Vol.1 No.2, pp.146 - 158

Published online: 28 Feb 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article