A Dynamic Virtual Network Function Orchestration Method Based on Deep Reinforcement Learning

Abstract

The invention relates to a dynamic virtual network function arrangement method based on deep reinforcement learning, which belongs to the field of wireless communication. The method includes: aiming at the high cost of VNF orchestration caused by dynamic changes in the physical network topology, establishing a mathematical model that minimizes the resource cost and operating cost of VNF orchestration under delay constraints; according to the dynamic changes in network topology and VNF dynamic changes, establish MDP model, and MDP is solved by deep Q network; Aiming at the problem of excessive state space and action space and dynamic change of network load in MDP model, a dynamic optimal VNF orchestration strategy is designed to solve the problem of high VNF orchestration cost. On the premise of ensuring the user delay performance, the present invention is limited by the computing resource capacity and link bandwidth resource capacity in the network, and dynamically adjusts the orchestration strategy of each network slice VNF to ensure user performance, optimize VNF orchestration costs, and improve resource utilization. utilization rate.

Description

technical field [0001] The invention belongs to the field of wireless communication, and relates to a dynamic virtual network function arrangement method based on deep reinforcement learning. Background technique [0002] For network service providers, providing services to users in an economical, green and efficient manner is a common pain point, because deploying services at the hardware level requires more time, cost and consumes more resources. The reliance on customized hardware has seriously hindered the development of the modern communication industry. The modern network industry highly requires lightweight service provision methods to promote network innovation and drive long-term expenditure reduction. The emergence of network function virtualization (Network Function Virtualization, NFV) technology will be the key to solving the above problems. Network function virtualization refers to the separation of network functions (such as firewalls, routers, etc.) , VNF)....

AI Technical Summary

Problems solved by technology

[0007] S1: Aiming at the high cost of virtual network function (Virtual Network Functional, VNF) orchestration caused by dynamic changes in physical network topology, establish a mathematical model that minimizes the resource cost and operating cost of VNF orchestration under delay constraints;

Images