The Optimized Deployment of Service Function Chain Based on Deep Reinforcement Learning Algorithm

초록

With the rapidly developing of mobile Internet market, although SDN and NFV technologies have enhanced the flexibility and scalability of the network, a large number of communication services still pose a great challenge to the allocation of network resources. In order to solve the problems of wasted resources and long latency in the deployment of Service Function Chain (SFC), this paper proposes the optimized deployment of Service Function Chain based on deep reinforcement learning. In this paper, we take minimizing the deployment cost as the objective, while considering the limitations of nodes, links, delay, and other resources. Then, we establish the corresponding optimization model and combine the deep reinforcement learning algorithm and particle swarm optimization to address the problem. Experimental results reveal that the proposed algorithm effectively converges faster and reduces the time delay of SFC deployment.

키워드