TCSA: A Dynamic Job Scheduling Algorithm for Computational Grids

Despite the popularity of parallel computing and the existence of distributed computing for several years, Grid computing, has adopted the ongoing demand for new and better ways to compute. In computational Grids, managing users' jobs and assigning them to appropriate resources become the first and foremost activities. In order to efficiently execute users' jobs and utilize the distributed resources in an appropriate way, this paper presents an effective job scheduling algorithm, called two choices scheduling algorithm (TCSA), to dynamically allocate jobs to resources so as to minimize the job execution time and maximize resource utilization. In this algorithm, instead of selecting just one of the available nodes to carry out a user submitted job, the allocation decided after examining a small number of the available nodes sampled randomly. To evaluate the performance of the proposed method, this study compares TCSA with those of several existing scheduling algorithms by conducting several simulation experiments under different scenarios. The obtained results show that the proposed TCSA algorithm can effectively reduce the makespan as well as balancing the load among resources compared with the other scheduling algorithms evaluated in this paper.