Application of Particle Swarm Optimization Algorithm based on Classification Strategies to Grid Task Scheduling

Grid task scheduling is a NP-hard problem. In this paper, an optimization algorithm of grid task scheduling is brought forward by using classification strategies to improve particle swarm algorithm. The particle swarm is divided into accurate subgroups for local slow search, commonness subgroups for the cloning strategy processing and inferior subgroups for changing into accurate subgroups to operate the positive and reverse clouds. The experimental results show that the scheduling algorithm effectively achieves the load balancing of resources and preferably avoids falling into local optimal solution and the selection pressure of genetic algorithm and elementary particle swarm algorithm. This algorithm has the high accuracy and convergence speed and so on.