An efficient performance evaluation model for the resource clusters in cloud environment using continuous time Markov chain and Poisson process

Cloud computing enables us to share the jobs across various cloud data centers with the internet as a backbone in order to process the jobs. The selections of the data centers to process the jobs are based on the potential of the data centers, arrival rate of the jobs and better resource utilization. Performance modeling and evaluation of the cloud systems shows a perfect picture of probability of distribution of jobs among the datacenters, mean number of jobs in the datacenters and makespan of the cloud environment. In order to provide a best performance evaluation model which leads to a prosperous cloud environment the proposed approach calculates the arrival rate of jobs using Poisson process in which the arrival rate can be calculated for infinite time intervals and the mean number of jobs and also the resource utilization are modeled using continuous Markov chain. The problems related to the calculation of arrival rate and resource utilization model of the existing approaches are solved to enhance the job scheduling by minimizing the makespan of the cloud environment.