Placement of Deep Reinforcement Learning in Service Function Chain of Network Function Virtualization

Reinforcement learning is a goal-oriented algorithm. The core goal of the algorithm is to learn to make decisions during the interaction between intelligent agents and the environment. Deep Reinforcement Learning is an algorithm that combines the perception ability of deep learning and the powerful decision-making ability of reinforcement learning. It can directly understand the network and task from experience, thus improving the accuracy and reducing repeated modeling, and is widely used to solve complex problems. A Network Service in NFV can be represented as a Service Function Chain (SFC), which consists of a series of Virtual network functions (VNF) in a given order. In order to provide efficient services, SFCS are deployed in a number of different servers and direct traffic through a set of VNFS in a specific order. This paper proposes SFC placement in NFV based on deep reinforcement learning model. Finally, it is found that the deep learning model can effectively improve the efficiency of SFC placement.