Abstract:Network function virtualization (NFV) makes the realization of specific network functions no longer depend on inherent hardware by executing virtual network functions (VNFs), but realizes network functions in a more flexible programming manner, thereby reducing the pressure of resource allocation on the underlying network. Service function chain (SFC) is composed of a set of fixed order VNFs. These VNFs need to be deployed on appropriate physical nodes to meet user function requirements, i.e., the placement of SFC. Traditional solutions mostly use mathematical models or heuristic methods, which are not applicable in the context of large-scale networks. Secondly, the existing methods do not integrate intelligent learning algorithms into the service function chain placement (SFCP) problem, which limits the possibility of obtaining better solutions. This paper presents a multi-objective optimization service function chain placement (MOO-SFCP) algorithm based on reinforcement learning (RL). The goal of the algorithm is to optimize the resource allocation mode, including several performance indexes such as underlying resource consumption revenue, revenue cost ratio, VNF acceptance rate and network latency. We model the SFCP as a Markov decision process (MDP), and use a two-layer policy network as an intelligent agent. In the training stage of RL, the agent comprehensively considers the optimization objectives and formulates the optimal physical node mapping strategy for VNF requests. In the test phase, the whole SFCP is completed according to the trained node mapping strategy. Simulation results show that the algorithm proposed in this paper has excellent performance in the aspects of underlying resource allocation revenue, VNF acceptance rate and so on. In addition, we prove that the algorithm has good flexibility by changing the delay constraint.

Efficient instance reuse approach for service function chain placement in mobile edge computing

Scheduling Reconfigurable Service Function Chain

Efficient Service Reconfiguration with Partial Virtual Network Function Migration

An online service function chain orchestration method for profit maximization in edge computing networks

Service Function Chain Placement for Joint Cost and Latency Optimization.

Efficient Dynamic Service Function Chain Combination of Network Function Virtualization

On the Effective Parallelization and Near-Optimal Deployment of Service Function Chains

Online VNF Chaining and Predictive Scheduling: Optimality and Trade-offs

Energy-aware Service Function Chaining Embedding in NFV Networks

A Multi-Agent Reinforcement Learning Scheme for SFC Placement in Edge Computing Networks

Virtual network function placement and resource optimization in NFV and edge computing enabled networks.

Service function chain network planning through offline, online and infeasibility restoration techniques

On Dynamic Service Function Chain Deployment and Readjustment

Efficiently Embedding Service Function Chains with Dynamic Virtual Network Function Placement in Geo-Distributed Cloud System

Towards deploying SFC with parallelized VNFs under resource demand uncertainty in mobile edge computing

Optimizing Service Function Chain Mapping in Network Function Virtualization through Simultaneous NF Decomposition and VNF Placement

Octans: Optimal Placement of Service Function Chains in Many-Core Systems

An Efficient Service Function Chaining Placement Algorithm in Mobile Edge Computing

Multi-objective Optimization Service Function Chain Placement Algorithm Based on Reinforcement Learning

Resource-Aware Service Function Chain Deployment in Cloud-Edge Environment

Composing and deploying parallelized service function chains