Optimization of Job Scheduling in the Cloud Computing Environment Using the Fuzzy Particle Swarm Optimization Algorithm

Nowadays, along with the constant increase of using cloud environment by companies and organizations, scheduling jobs in this environment in an optimum way is of prime importance. Therefore, different algorithms have been suggested for assigning tasks to resources in cloud environments; however, most of which do not consider criteria such as balanced load, and reduction of the task completion time. In this work, using the meta-heuristic algorithm of swarm particles optimization (PSO) and fuzzy logic, task completion time is reduced, and, as a result of which, efficiency of using resources is increased. Generally, in a distributed system like cloud environment, tasks are assigned randomly to resources. Hence, total load on the cloud environment could become imbalanced, which reduces system’s efficiency. In this research, PSO and fuzzy logic is used for job scheduling. In addition, the use of simulated annealing (SA) to improve the initial solutions, which are generated randomly, is suggested. Results show that the suggested optimization method can effectively improve criteria like makespan once compared with results of algorithms without optimization, like Ron-robin, and even in comparison to other optimization algorithms, like genetic algorithm.