Multi-objective Workflow Grid Scheduling Based on Discrete Particle Swarm Optimization
Grid computing infrastructure emerged as a next generation of high performance computing by providing availability of vast heterogenous resources. In the dynamic envirnment of grid, a schedling decision is still challenging. In this paper, we present efficient scheduling scheme for workflow grid based on discrete particle swarm optimization. We attempt to create an optimized schedule by considering two conflicting objectives, namely the execution time (makespan) and total cost, for workflow execution. Multiple solutions have been produced using non dominated sort particle swarm optimization (NSPSO) . Moreover, the selection of a solution out of multiple solutions has been left to the user. The effectiveness of the used algorithm is demostrated by comparing it with well known genetic algorithm NSGA-II. Simulation analysis manifests that NSPSO is able to find set of optimal solutions with better convergence and uniform diversity in small computation overhead.
KeywordsParticle Swarm Optimization Pareto Optimal Solution Pareto Optimal Front Discrete Particle Swarm Optimization Conflicting Objective
Unable to display preview. Download preview PDF.
- 1.Abraham, A., Liu, H., Zhang, W., Chang, T.G.: Scheduling Jobs on Computational Grids Using Fuzzy Particle Swarm Algorithm, pp. 500–507. Springer, Heidelberg (2006)Google Scholar
- 4.Buyya, R., Venugopal, S.: A Gentle Introduction to Grid Computing and Technologies. CSI Communications 29, 9–19 (2005)Google Scholar
- 5.Buyya, R., Murshed, M.: GridSim: A Toolkit for Modeling and Simulation of Grid Resource Management and Scheduling, vol. 14, pp. 1175–1220 (2002), http://www.buyya.com/gridsim
- 8.Deb, K., Jain, S.: Running Performance Metrics for Evolutionary Multi-objective Optimization. In: Proceedings of Simulated Evolution and Learning (SEAL 2002), pp. 13–20 (2002)Google Scholar
- 12.Kennedy, J., Eberhart, R.: Particle Swarm Optimization. In: Proceedings of IEEE International Conference on Neural Networks, vol. 4, pp. 1942–1948 (1995)Google Scholar
- 13.Li, X.: A Non-dominated Sorting Particle Swarm Optimizer for Multi- objective Optimization. In: Proceeding of Genetic and Evolutionary Computation Conference 2003 (GECCO 20003), Chicago, USA (2003)Google Scholar
- 14.Ritchie, G., Levine, J.: A fast, effective local search for scheduling independent jobs in heterogeneous computing environments, Technical report, Centre for Intelligent Systems and their Applications, School of Informatics, University of Edinburgh (2003)Google Scholar
- 16.Tsiakkouri, E., Sakellariou, R., Zhao, H., Dikaiakos, M.D.: Scheduling Workflows with Budget Constraints. In: CoreGRID Integration Workshop Pisa, Italy (2005)Google Scholar
- 19.Praveen, K., Das, S., Welch, S.M.: Multi-Objective Hybrid PSO Using ε-Fuzzy Dominance. In: Proceeding of Genetic and Evolutionary Computation Conference (GECCO 2007), London, UK (2007)Google Scholar