Game-Based Negotiation between Power Demand and Supply in Green Datacenters

International audienceThe power consumption of datacenters is growingrapidly and becoming a major concern. For reducing carbonfootprint and increasing energy efficiency, a promising solutionis to locally supply datacenters with renewable energies. However,a challenging problem in building such green datacenter is thecoordinating between the power demand and the intermittentpower supply. To address this problem, we propose to modelthe green datacenter as two subsystems, namely, InformationTechnology (IT) subsystem which consumes energy, and electricalsubsystem which supplies energy. Then we aim to find an efficientcompromise between the power supply and power demand,taking into account the constraints of both subsystems. Basedon buyer-supplier game, we introduce a negotiation approach, inwhich the two subsystems are modeled as the energy buyer andenergy supplier. A negotiation algorithm is proposed, allowingthe two subsystems to negotiate and reach an efficient trade-off,while respecting their ownutility/monetary gain. The algorithmis evaluated in our middleware of renewable energies-powereddatacenter. The experimental results show that the proposedalgorithm allows the negotiation process to reach stable points.This algorithm also obtains significant improvement in thedatacenter’s utility and quality of service (QoS), compared tothe algorithms in which joint IT-energy management is notconsidered