Services utility prediction on a cloud

Cloud computing and Software-as-a-Service (SaaS) are being widely applied. In the near future, there will be many cloud services providing similar services. One of the issues is to dispatch service requests to available services dynamically so that service requests can be served in the most effective and efficient way. This paper proposes an algorithm that efficiently serves requests for services by predicting the number of services needed. Furthermore, a cloud architectural framework to support the algorithm and prediction model is introduced. The paper also presents a simulation study to validate the algorithm proposed.