An Online Cost-Based Job Scheduling Method by Cellular Automata in Cloud Computing Environment

Abstract

Cloud computing has expanded considerably in industry and research and is based on a pay-as-you-go payment model. In cloud computing environment, on one hand, jobs sent to the cloud to execution have a variety of attribute such as deadline, length, bandwidth requirements. On the other hand, various virtual machines have been created at different costs on existing physical resources. In this paper, a job scheduling method is proposed that carries out scheduling using cellular automata. The proposed algorithm is called CA-JS. The main goal of this method is to execute the jobs in the specified deadline and to increase the profitability of cloud providers. Also, in this paper, another attribute named hardness factor, is determined by each user of jobs sent to the cloud, which also specifies the running cost of jobs. The simulations carried out in the CloudSim environment indicate that the proposed method, in comparison with FCFS, Min–Min, and EDF algorithms, has better tardiness and makespan, and also, allows more jobs to be executed in their specified deadline.