A queueing-based model for performance management on cloud

Cloud computing is a new trend for computing resource provision. Many public clouds are available for developers to transfer or build web applications on cloud. As a result, the computing resource scheduling and performance managing have been ones of the most important aspects of clouding computing. In this paper, we propose a queueing-based model for performance management on cloud. In this model, the web applications are modeled as queues and the virtual machines are modeled as service centers. We apply the queueing theory onto how to dynamically create and remove virtual machines in order to implement scaling up and down. There is no VM (Virtual Machine) live migration involved in this model which makes it much simpler than some existing models. The result of case study has shown this model is effective to scaling up and down. The more precise measurement and analysis of this model will be done in future.