Rank Based Genetic Scheduler for Grid Computing Systems

Computational grids have become attractive and promising platforms for solving large-scale high-performance applications of multi-institutional interest. However, the management of resources and computational tasks is a critical and complex undertaking as these resources and tasks are geographically distributed and a heterogeneous in nature. This paper proposes a novel Rank Based Genetic Scheduler for Grid Computing Systems (RGSGCS) for scheduling independent tasks in the grid environment by minimizing Make span and Flow time. The novel RGSGCS speeds up convergence and shortens the search time better than Standard Genetic Algorithm (SGA) using Rank-based fitness, at the same time the heuristic initialization of initial population using Minimum Completion Time (MCT) heuristic which allows RGSGCS to obtain a high quality feasible scheduling solution. The simulation results show that RGSGCS has better search time than SGA.