Particle swarm optimization for scheduling problems by curve controlling based global communication topology

Execution of workflow application in cloud computing is widely applied, it is to apply heterogeneous resource in cloud for meeting the demand for large computational requirements. To maximize the efficiency of cloud computing, we have to effectively solve resources arrangement problem of workflow application in cloud computing, namely efficient task scheduling is needed. Cloud task scheduling problem is an NP-hard problem; therefore, a modified particle swarm optimization (PSO) is used to solve the task scheduling problem in the cloud, the grid task scheduling problems including how to distribute tasks to cloud resource and how to place a priority of tasks. In this study, two types of PSO are applied to solve these two issues, discrete particle swarm optimization is used to solve problem of distribution tasks in the cloud and a modified conventional PSO, named global topology controlling PSO (GTCPSO), is used in task priority assignments. In order to increase PSO searching efficiency, curve controlling is used to adjust the usage of global topology communication between particles so as to control search behavior of particle swarm. Thus, the swarm can perform global search at the initial stage, and then does fully local search at the last stage. The experiment results indicate that the curve controlling of topology communication between particles based on a designed global topology rate can efficiently solve the task scheduling problem and enhance performance of cloud computing.