Non Dominated Particle Swarm Optimization For Scheduling Independent Tasks On Heterogeneous Distributed Environments

Scheduling tasks is one of the core steps to effectively exploit the capabilities of distributed or parallel computing systems. In general, scheduling is an NP-hard problem. Most existing approaches for scheduling deal with a single objective only. This paper presents a multi-objective scheduling algorithm based on particle swarm optimization (PSO). In this paper a non-dominated sorting particle swarm optimization (NSPSO) that combines the operations of NSGA –II is used to schedule tasks in a heterogeneous environment. The approach aims at developing optimal schedules thereby minimizing two objectives, makespan and flowtime simultaneously. The experimental results indicate that NSPSO obtains good solutions on benchmark instances in comparison with a multi-objective particle swarm optimizer using a weighted approach (W-MOPSO )which is also implemented in this paper for effective comparisons.