Journals Proceedings

Abstract

Scheduling is the main difficulty in heterogeneous computing (HC) systems in achieving the high performance. In this paper, a meta-heuristic approach based on Particle Swarm Optimization (PSO) is adopted for solving task scheduling problem. PSO is a population-based algorithm to find the optimal solutions, but its performance is decreased when considering multi-optimization problem. In this paper, an Adaptive Weighted Particle Swarm Optimization is proposed for multi-objective optimization. AWPSO is an efficient and simple tool for multi-objective and multi-dimensional problem. AWPSO enhance the global search ability and to overcome the local optimum by introducing an acceleration factor. The goal is to minimize the makespan and flowtime. The experimental results showed that the performance of the proposed method is effective compared with other heuristic optimization technique namely PSO in finding the optimal solutions.