An Approach to Reduce Energy Costs Through Virtual Machine Migrations in Cloud Federation

摘要

Cloud federation offers new business models to enforce more flexible energy management strategies. Independent Cloud providers are exclusively bounded to the specific energy supplier powering its Data Centers. The situation radically change if we consider a federation of cooperating Cloud providers. In such a context a proper migration of virtual machines among providers can lead to a global energy cost-saving strategy. In this paper, we present an approach to reduce energy cost in a federated Cloud ecosystem. More specifically, we propose an algorithm that allows providers to determine a map of possible destinations for cost-evaluation. Furthermore, we introduce an additional algorithm to determine the optimum energy cost migration path, and, consequently, the best Cloud Data Center where virtual machines should be migrated in order to push down energy costs.