The SAGITTA Approach for Optimizing Solar Energy Consumption in Distributed Clouds with Stochastic Modeling

Abstract

Facing the urgent need to decrease data centers' energy consumption, Cloud providers resort to on-site renewable energy production. Solar energy can thus be used to power data centers. Yet this energy production is intrinsically fluctuating over time and depending on the geographical location. In this paper, we propose a stochastic modeling for optimizing solar energy consumption in distributed clouds. Our approach, named SAGITTA (Stochastic Approach for Green consumption In disTributed daTA centers), is shown to produce a virtual machine scheduling close to the optimal algorithm in terms of energy savings and to outperform classical round-robin approaches over varying Cloud workloads and real solar energy generation traces.

Integrating Pre-Cooling of Data Center operated with Renewable Energies

“…They studied 42 locations to assess their production and free cooling potential and gave results on optimal sizing for hybrid power infrastructure. Other works study geographical allocation of tasks on data center having their own on-site or off-site renewable sources [11]- [13], [23].…”