Research on Service Function Chain Deployment Algorithm Based on Proximal Policy Optimization

In the future, business scenarios will become diversified, but it is difficult for the existing network architecture to provide strong support for them. Network function virtualization (NFV) technology decouples network functions from dedicated hardware devices and provides customized services for users in the form of service function chain (SFC). At present, the deployment of SFC has been proved to be a NP-hard problem. Most of the solutions are integer linear programming algorithms, but the process of such algorithms is complex. When the network topology scale becomes larger, the calculation process is very time-consuming, and the results sometimes fall into local optimal solutions, which makes it difficult to achieve the desired effect. In this case, reinforcement learning (RL) algorithms show great advantages, learning strategies through interaction with the environment to maximize rewards or achieve specific goals. Therefore, this paper proposes a SFC deployment algorithm based on proximal policy optimization (PPO) reinforcement learning, which aims at maximizing access rate and minimizing resource consumption. The simulation results show that the proposed algorithm has good convergence and stability, which is more conducive to the actual deployment of the SFC.