Dual-level cooperative game approach for energy-aware resource allocation in data centers

Abstract

Cloud computing is the next major paradigm shift in information and communication technology. Today, data center (DC) is pivotal to the effective implementation of the cloud computing paradigm. To deliver the necessary cloud services, the DC faces major reliability and robustness challenges. In this study, we concentrate upon the efficient DC resource allocation technique by considering the synergy of the performance and energy efficiency. To achieve the maximum system efficiency, thermal-aware control mechanism is developed based on the dual-level cooperative game. Proactively, we adaptively assign the workload among different servers using the concept of relative utilitarian bargaining solution. Reactively, we maintain a thermal equilibrium according to the characteristics of τ-value. The main advantage of our dual-level game model is to provide an axiom-based strategic solution for the DC resource allocation problem while dynamically responding to the current workload and thermal conditions. The effectiveness of our proposed protocol is verified through simulation evaluations under different workload scenarios. Results demonstrate that the proposed approach significantly outperforms the existing state-of-the-art DC control schemes from energy consumption and load balancing perspectives. Last, some open problems are identified and we summarize possible research issues for the future study.