부산대학교 도서관

Container and microservice-based resource management approach can achieve the processing and computation of massive and highly concurrent service data in distribution station area. However, resource management in distribution station area still faces the challenge of large data computation delay and inability to achieve service priority awareness. In this paper, a container and microservice-based resource management architecture is constructed, and an upper confidence bound (UCB)-based resource management capability optimization algorithm is proposed to minimize data computation delay. The algorithm can resolve the conflicts between differentiated microservices and achieve more rational exploration by considering microservice priorities to improve resource utilization. Simulation results show that the proposed algorithm has superior performance in data computation delay with differentiated microservice priorities.