Application of A3C Algorithm in Dynamic and Parallel SFC Placement Framework

NFV realizes the sharing of network resources by changing the deployment form of traditional networks, and provides technical support for the realization of dynamic functional service chain. In NFV environment, users' service requests are fulfilled by SFC. The SFC orchestration problem includes SFC deployment and reconfiguration. An SFC is an end-to-end virtual network composed of several service-related VNF connected in order. The current SFC deployment and resource allocation methods may not be sufficient to support the performance requirements of SFC. In this case, it is necessary to reconfigure SFC to maintain normal service performance. The A3C algorithm can asynchronously multithread the process of interactive learning between multiple agents and the environment at the same time. This paper presents application of A3C algorithm in dynamic and parallel SFC placement framework. Finally, the experiment of the A3C algorithm is carried out, and the experiment proves that the proposed algorithm can effectively improve the efficiency.