Abstract: The Cloud computing is one of the emerging modern technologies that need to meet the emerging business needs for agility, flexibility, cost reduction and time-to-value. The developments in cloud computing paradigm necessitate faster and efficient performance evaluation of cloud computing servers. The advanced modeling of Cloud servers are not feasible one due to nature of cloud servers and diversity of user requests. This issue necessitates calculating performance of cloud servers and should be capable of handling several millions of requests in few seconds. It is so possible through the specialized computing facility called Virtualization and even with it, very hard to serve plenty of requests continuously or simultaneously without delay or collision. We proposed a new mechanism to overcome these performance problems and to aim for optimize their performance; we can achieve this goal with help of queuing theory. To tackle the problem of optimizing the performance of cloud servers, we model the cloud servers with multitenant architecture as an M/G/m/m + r queuing system with single task arrivals and a task buffer of finite capacity. In this model focusing on input buffer size, number of servers, mean number of request, response and etc. as metrics for performance. The proposed algorithm named approximate Analytical model, and is equivalent to the combination of Transformation based Analytical Model & an Approximate Markov Chain Model with supporting calculations works out to bring the optimized performance evaluation of cloud servers.
Keywords: Cloud Computing, Virtualization, and performance, Queuing theory.
Title: OPTIMIZED PERFORMANCE EVALUATIONS OF CLOUD COMPUTING SERVERS
Author: K. Sarathkumar
International Journal of Computer Science and Information Technology Research
ISSN 2348-120X (online), ISSN 2348-1196 (print)
Research Publish Journals
