Branching Processes Theory Application for Cloud Computing Demand Modeling Based on Traffic Prediction
Session Details
Session Abstract
The potential benefits of cloud computing are overwhelming. However, attaining these benefits requires that each aspect of cloud environment support the key design principles of the cloud model. One of the core design principles is dynamic scalability, or the ability to provision and decommission servers on demand. The efficient data processing is a fundamental and vital issue for almost every business organization. Unfortunately, the majority of today’s database servers are incapable of satisfying this requirement. From the viewpoint of a service provider, demands on the network are not entirely predictable. 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. Branching stochastic processes are used to describe random systems such as nuclear chain reactions, decommission population development, epidemic of disease and gene propagation. In this paper we show that the cloud computing demand can be developed as a branching stochastic process. A statistical model is described step by step, as a function of the physical parameters of the process. Using the model, we propose a method for determining the unknown probability distribution of queries.