Branching Processes Theory Application for Cloud Computing Demand Modeling Based on Traffic Prediction

With cloud computing growing in popularity cloud-service providers must guarantee that data are processed rapidly and transferred when and where they are needed. Unfortunately, it is extremely difficult to predict the exact performance characteristics and demands on the network at any particular time. In this paper we show that the cloud computing demand can be developed as a branching stochastic process. Branching processes are used to describe random systems such as population development, nuclear chain reactions and spread of epidemic disease. A statistical model is described and using this model we propose a method for determining the unknown probability distribution of queries. Network traffic modeling is an issue of great importance to both consumers and providers of cloud-based services. Firstly, traffic modeling helps to represent our understanding of dynamic demand for cloud services by stochastic processes. Secondly, accurate traffic models are necessary for service providers to properly maintain quality of service.