Optimizing Energy Efficiency in the Cloud Using Service Composition and Runtime Adaptation Techniques

This paper describes an approach for service composition optimization and its application in cloud computing to streamline resource usage that in turn contributes towards energy efficiency. The suitability and usefulness of the approach is evaluated by experimentation. In the experiments, physical hosts at various cloud sites represent candidate services that are brought together in a composition to satisfy the requirements of applications. The composition is optimized based on functional and non-functional criteria to determine a set of cloud services representing energy efficient deployment configurations. We also propose a runtime adaptation model that can help in minimizing energy consumption of cloud applications at runtime.