Abstarct: Increasing the number of data centers around the world, their owners have been faced with the problem of high operating costs. The cost management of power consumption required for data centers has been recently paid more attention. In this thesis, we study a system that involves distributed data centers across different geographic regions under different day-ahead electricity markets. Each data center is equipped with high-capacity energy storage system. The whole system is programmed in such a way that the cost of power consumption reduces by optimal workload dispatching and optimal energy management of storage systems. The system is modeled regarding power balancing constraints, battery cost and QoS. For more practical results a penalty function is also considered in case QoS constraints are not met. Furthermore, for more accuracy the model has been extended to consider the effects of batteries' depth of discharge in the cost of energy storage. Also the problem have been linearized for comfortable solving and simulated to optimize the cost of power consumption. The results show that not only the scheme of power consumption is improved, but also the cost of power consumption is also decreased by 32% and achieve an optimal trade-off between battery cost and electricity cost saving.