Performance Modeling for a Cloud Computing Center Using GE/G/m/k Queuing System

Cloud computing is a new paradigm for the provision of computing infrastructure, it aims to change the location of the computer on the network infrastructure to reduce the maintenance costs of hardware and software resources. Increasing demand in cloud computing, in recent time, pushes researchers to improve access to services by developing techniques that allow users in cloud computing networks to get better services in terms of optimum performances with lowest possible cost. To evaluate the performances of a cloud computing center, most related prior studies were motivated by using analytical models based on queuing theory. However, and to the best of our knowledge, a model taking into account the assumption of batch arrivals in the center has never been studied before. In this paper, this case is studied by modeling the cloud data center as a (GE/G/m/k) queuing system with GE distribution task arrivals, a general service time for requests as well as large number of physical servers and a task buffer of finite capacity. We used this model to evaluate the performance analysis of cloud server frames and we get an accurate estimate of the complete probability distribution of the request response time and other important performance indicators such as: the mean number of tasks in the system, the distribution of waiting time, the probability of immediate service and the blocking probability.