Abstract: The growth of energy costs and the increasing environmental problems make the data center face severe challenges,and the introduction of economic environment-friendly new energy is imminent.But the new energy has several cha-racteristics such as intermittent,unstable and mutagenic,causing the data center can not effectively adapt new energy.To this end,the major data centers put forward many solutions,such as energy management strategies and load scheduling algorithms,etc.However,most of the existing research results emphasize the optimization of the calculation of energy consumption,and it can not be further adapted to storage.Therefore,this paper presented an energy consumption optimization scheme for new energy-driven storage system,and used the characteristics of different storage media and online-offline workload classification model to implement the matching between workload energy comsumption requirement and new energy supplying.In order to guarantee the performance and improve the energy efficiency of the storage system,the dual-drive energy control method and the virtualization consolidation technology were used to achieve fine-grained energy control program.In addition,this paper also designed and implemented an offline-workload scheduling algorithm to further improve the utilization of new energy.The experimental results show that the optimization scheme for new energy-driven storage system can make the utilization rate of new energy reach 95%,while ensuring the degradation rate of the storage system is less than 9.8%.

Key words: Energy optimization, Heterogeneous storage systems, New energy-driven, Virtualization consolidation, Workload scheduling