Greening Data Centers within Cost Budget

Along with the increasing Internet services and cloud computing in recent years, the power usage associated with data center operations has been surging significantly. This huge energy consumption leads to not only heavy operation cost but also negative environmental impact. To reduce the carbon footprint, many cloud service providers have announced plans to power their data centers with renewable energy. However, current renewable energy is usually more expensive than traditional brown energy. Using renewable energy may further increase the operation cost and impose a heavy burden on the cloud service providers. Therefore, it´s important to increase the renewable energy utilization within a long-term (e.g., a month) cost budget. In this paper, we investigate how to utilize the temporal diversities of electricity price and green energy availability to green data center within the specified cost budget. To this end, we model the long-term cost budget constraint and formulate the renewable energy utilization maximization problem as a stochastic programming problem. While all future knowledge of electricity price, workload and renewable energy availability are unknown, we apply Lyapunov optimization framework and design an efficient online algorithm. Extensive performance evaluations show that the proposed algorithm can significantly increase the renewable energy utilization within the cost budget.