Performance evaluation of dynamic resource allocation in cloud computing platforms using Stochastic Process Algebra

Abstract

Cloud computing is a revolutionary concept in which computing services are provided to end users via virtualization. Computing resource are provisioned and released on demand. This elastic nature of resource provisioning allows for the “pay as you go” concept. Hence, it makes the major benefit of adopting cloud-based solutions. Accordingly, there is a need to evaluate the performance of the resource provisioning algorithm adopted by cloud computing platforms. In this paper, an approach for modeling and analyzing the performance of the resource provisioning process in a cloud computing platform is presented. The proposed model utilizes Stochastic Process Algebra (SPA) as an engine to simulate the components involved in the resource provisioning process along with the interactions among them. A case study using the cloud computing platform Aneka is performed. Experimental results show the effect of different parameters on the provisioning process which suggests the suitability of the proposed approach to evaluate design alternatives.