Service Function Chain Embedding for NFV-Enabled IoT Based on Deep Reinforcement Learning

Abstract

It is challenging to efficiently manage different resources in the IoT. Recently, Network function virtualization has attracted attention because of its prospect to achieve efficient resource management for IoT. In NFV-enabled IoT infrastructure, a service function chain (SFC) is composed of an ordered set of virtual network functions (VNFs) that are connected based on the business logic of service providers. However, the inefficiency of the SFC embedding process is one major problem due to the dynamic nature of IoT networks and the abundance of IoT terminals. In this article, we decompose the complex VNFs into smaller VNF components (VNFCs) to make more effective decisions since VNF nodes and physical network devices are usually heterogeneous. In addition, a deep reinforcement learning (DRL)-based scheme with experience replay and target network is proposed as a solution that can efficiently handle complex and dynamic SFC embedding scenarios. Simulation results present the efficient performance of the proposed DRL-based dynamic SFC embedding scheme.