The rapid growth of cloud computing has exacerbated data center energy consumption issues,necessitating intelligent and efficient optimization methods.This paper analyzes the problem,explores the potential of reinforcement learning in energy scheduling,constructs a comprehensive data center model,formalizes the scheduling problem as a Markov decision process,and proposes a deep reinforcement learning algorithm combining graph neural networks and long short-term memory networks.Simulations verify the effectiveness of the proposed strategy.
data centerenergy optimizationreinforcement learningscheduling
NETL
NSTL
万方数据
