Reducing Energy Consumption for Reconfiguration in Cloud Data Centers

Mobile Cloud Computing (MCC) leverages mobile devices and infrastructure equipment to increase services accessibility. It uses increased devices computing capability to enhance services usability and ensure high availability. This growth in performances results in an increased interest for platforms use to accommodate a multitude of applications. To support such an increase in demand, new designs for resource management have to be implemented in order to reach usage optimality. In this work, we propose to design new algorithms to optimize MCC resources management techniques based on stochastic networks optimization. Our approach is focused on energy consumption optimization on the cloud data center side while ensuring resources elasticity to adapt to users' demands and insure a highly available platform. We elected an overclocking technique to enhance servers' capabilities and Lyapunov improvisation to ensure design stability and to minimize the energy cost. We perform extensive simulations under different charge conditions in order to prove the design effectiveness in ensuring the service with lower power consumption. Simulations results confirm the effectiveness of the proposed resources management design.