A Discrete Particle Swarm Optimization Approach for Grid Job Scheduling

Scheduling is one of the core steps to efficiently exploit the capabilities of emergent computational systems such as grid. Grid environment is a dynamic, heterogeneous and unpredictable one sharing different services among many different users. Because of heterogeneous and dynamic nature of grid, the methods used in traditional systems could not be applied to grid scheduling and therefore new methods should be looked for. This paper represents a discrete Particle Swarm Optimization (DPSO) approach for grid job scheduling. PSO is a population-based search algorithm based on the simulation of the social behavior of bird flocking and fish schooling. Particles fly in problem search space to find optimal or near-optimal solutions. In this paper, the scheduler aims at minimizing makespan and flowtime simultaneously in grid environment. Experimental studies illustrate that the proposed method is more efficient and surpasses those of reported meta-heuristic algorithms for this problem.