Distributed green data center energy management with combined cooling system
In response to the construction of green data centers for sustainable development,reducing data center power consumption and carbon dioxide emissions has become an urgent issue.A data center model considering cooling systems was established.The model transforms cost functions of data center power costs,bandwidth costs,carbon emission costs,and cooling costs into a cost minimization problem.The data center operational costs are reduced by effective workload scheduling and rational servers on-off strategies.The electricity prices and carbon emission rates are difficult to predict accurately due to the fluctuation.A feature of the Lyapunov optimization framework was utilized to propose an online control strategy.The strategy only requires knowledge of the current system information.An alternating direction method of multipliers(ADMM)is introduced to enable the control center to coordinate workload distribution among various data centers.Data centers only need to exchange workload decisions,effectively reducing computational complexity and safeguarding user privacy.Simulation results based on real-world data demonstrate that the proposed algorithms can effectively reduce data center operational costs.
distributed data centerenergy efficiency managementcost optimizationLyapunov optimization
NETL
NSTL
万方数据
