Optimized performance measurements of cloud computing center

Abstract

The emergence of various technologies results in the increase in number of cloud users. People are themselves associated with the cloud environment with or without their knowledge. The cloud service providers are forced to establish and manage various data centers for providing services to the end users in a desirable way. It has been noticed that there exists a similarity between the cloud environment and the queuing models. So it is proposed to design the cloud environment as a queue model and analyze the various performance parameters of the cloud data centre. Two queuing models were used M/M/m queuing model and M/G/m/m+r. The cloud environment is designed as M/M/m queuing model and a single server is used, hence the queue model becomes M/M/1 queuing model. The performance parameters that are considered are response time, waiting time and utilization of the CPU. While the cloud environment is modeled as M/G/m/m+r queuing model, the system is tested with different number of servers. The discipline that is used with both the models is first come first serve model. The objective of the proposed work is to analyze the performance of the data centre. The parameters that are measured are response time, waiting time and the utilization of the central processing unit for different arrival rates. These performance parameters are measured with the help of a tool called oracleand#8223;s automatic workload repository report that has been integrated with the given model. With this the actual capacity of the individual servers and in turn the capacity of the data centre at which the desirable service can be provided to the end user can be measure newline