Data center optimization methodology to maximize the usage of locally produced renewable energy

In this paper we address the problem of Data Centers energy efficiency by proposing a methodology which aims at planning the Data Center operation such that the usage of locally produced renewable energy is maximized. We defined a flexibility mechanism and model for Data Center's components (electrical cooling system, IT workload, energy storage and diesel generators) leveraging on optimization actions such as load time shifting, alternative usage of non-electrical cooling devices such as the thermal storage or charging/discharging the electrical storage devices, etc. The flexibility mechanism enacts the possibility of shifting the Data Center's energy demand profile from time intervals with limited renewable energy production due to weather conditions, to time intervals when spikes of renewable energy production are predicted. We have developed a simulation environment which allows the methodology to be inlab tested and evaluated. Results are promising showing an increase of renewable energy usage of 12% due to energy consumption demand shift for following the renewable energy production levels.