Cloud Computing Performance Evaluation under Dynamic Resource Utilization and Traffic Control
Mohamed Hanini, Said El Kafhali · 2017
Cloud computing has emerged as a new commercial infrastructure paradigm which has revolutionized the IT industry. It has particularly the advantage of delivering a flexible, high-performance, pay-as-you-go, on-demand service. Maintaining the Quality of Service (QoS) at an acceptable level for cloud customers requires an accurate and well adapted performance analysis mechanism. The QoS required by customers is guaranteed using a Service Level Agreement (SLA) document. This has increased the complexity for service providers as they now have to not only managed their resources but also they need to provide expected QoS for customers. In this paper, we study a model which associates a scheme for Virtual Machine (VM) instantiation in a given physical machine (PM) with a mechanism to control the access for incoming requests to the Virtual Machine Monitor (VMM). In this model the number of activated VMs in the PM is defined according to the number of jobs in the system, and the control access is based on the load of the system. We develop an analytical model, derive performance parameters expressions and give numerical examples to show how the model estimates these QoS parameters. In particular, we plot the loss probability, mean number of requests in the system and mean requests delay while varying the incoming request arrival rate for two systems (system under control and system with no control). We cross-validate our analytical model using MATLAB software implementation. Analysis results obtained show the positive impact of the proposed mechanism on the QoS parameters.