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.

Software Defined Resource Allocation for Service-Oriented Networks

Network Slicing for Service-Oriented Networks Under Resource Constraints

Hierarchical Resource Allocation in Multi-Service Wireless Networks with Wireless Network Virtualization

Scheduling Reconfigurable Service Function Chain

Delay-Aware Resource Management for Multi-Service Coexisting LTE-D2D Networks with Wireless Network Virtualization

Network Slicing for Service-Oriented Networks with Flexible Routing and Guaranteed E2E Latency

A Method for Adaptive Resource Adjustment of Dynamic Service Function Chain.

Service Function Chain Placement for Joint Cost and Latency Optimization.

Congestion-Aware Algorithms for Service Function Chaining in Software-Defined Wide Area Networks.

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

Resource Aware Routing for Service Function Chains in SDN and NFV-Enabled Network

Resource Management for Heterogeneous Services in Cognitive Mesh Networks

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

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

Service-Oriented Resource Allocation in Clouds: Pursuing Flexibility and Efficiency

Sdn Based Vxlan Optimization In Cloud Computing Networks

A Reliable Resource Scheduling for Network Function Virtualization.

Joint Network Function Placement and Routing Optimization in Dynamic Software-defined Satellite-Terrestrial Integrated Networks

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

Application-Driven Provisioning of Service Function Chains Over Heterogeneous NFV Platforms