Analysis of performance and energy consumption in the cloud

Abstract

We analyze here a cloud system represented by hysteresis multi server queueing system. It is characterized by forward and backward thresholds for activation and deactivation of block of servers representing a set of VMs (Virtual Machines). The system is represented by a complex Markov Chain which is difficult to analyse when the size of the system is huge. We propose both analytical and numerical mathematical methods for deriving the steady-state probability distribution. We compute then performance and energy consumption measures and we define an overall cost taking into account both aspects. We compare the proposed methods with respect to the computation time and we analyse the impact of some parameters on the behaviour of the system.