High Exploitation Genetic Algorithm for Job Scheduling on Grid Computing

Scheduling jobs on computational grids is identified as NP-hard problem due to the heterogeneity of resources; the resources belong to different administrative domains and apply different management policies. Genetic algorithm which is a metaheuristic search on the basis of the idea of the natural evolution of living organisms generate solutions in order to reach the best solution, using techniques inspired by nature, such as the selection, crossover and mutation. One of the most important processes in the genetic algorithm is the crossover process that combines two chromosomes (parents) to produce a new chromosome (offspring). The parents with the highest fitness functions are selected to participate in the process. The idea behind crossover is that the new chromosome will be better than both parents because it takes the best qualities of both of them. This paper proposed a new job scheduling mechanism based on increasing the crossover rate in genetic algorithm in order to reach the best solution faster to improve the functionality of the genetic algorithm. To evaluate the proposed mechanism this study conducted a simulation using GridSim simulator and different workloads. The results of the simulation process revealed that the increase in the exploitation process decrease the finish time.