MULTI-AGENT REINFORCEMENT LEARNING SCHEDULING METHOD AND SYSTEM AND ELECTRONIC DEVICE

摘要

Disclosed are a multi-agent reinforcement learning scheduling method and system and an electronic device. The method comprises: step a: collecting server parameters of a network data center and load information of virtual machines running on each server (100); step b: establishing a virtual simulation environment by using the server parameters and the load information of the virtual machines, and building a multi-agent deep reinforcement learning model; step c: performing offline training and learning by using the multi-agent deep reinforcement learning model, and training an agent model for each server; and step d: deploying the agent model to a real service node, and scheduling according to the load condition of each service node. The virtualization technology is used for virtualizing the services running on the server and the virtual machines are scheduled for load balancing, thereby achieving more macroscopic resource allocation and realizing the collaboration strategy of multi-agents in a complex dynamic environment.