Design of SLA-based VM Provisioning Mechanism for Cloud Computing

Abstract

Cloud computing is the latest computing paradigm that delivers computing resources as services. Achieving QoS (Quality of Service) requirements of cloud users is important for meeting SLA (Service Level Agreement) contracts. However, QoS requirements are difficult to satisfy due to the high variability of users' needs. To accommodate the variability efficiently, cloud providers need to predict users' needs and to estimate the amount of the resources prepared. In this paper, an optimization model of dynamic VM provisioning mechanism is presented to elastically estimate the amount of resource to be prepared by maximizing a provider's revenue while satisfying users' QoS requirements and minimizing the penalty for SLA violation. Experimental results show that our model elastically adapts to the fluctuation in resource demands while maximizing the revenue of a cloud provider. © 2015 International Information Institute. Author keywords