A hybrid batch job scheduling algorithm for grid environment

Grid computing is a collection of geographically heterogeneous distributed computational resources that enables users for sharing data and other computing resources. One of the major challenges in grid computing is how to schedule batch jobs across such an environment with minimum makespan (the finishing time of the last job) and flow time. In this study, a hybrid batch job scheduling method is proposed for grid environment that combines genetic and particle swarm optimization techniques to reduce makespan and flowtime. Experimental results show a reduction in makespan for 7 out of 12 instances of Braun workload comparing to minmin, maxmin, and discrete PSO algorithms.