Research On Technology Of Customizing NFV Network Services

In a Network Function Virtualization(NFV)environment,Service Function Chaining(SFC)customizes a certain network service by connecting an ordered set of Virtual Network Functions(VNF)which can be flexible to provide network services.However,at present,SFC only customizes network services by defining the execution order of VNFs without considering the operating parameters and timing of VNFs.It is difficult to meet the needs of flexibly customizing NFV network services and finely controlling their operation.The goal of this dissertation is to research and develop a technology that can flexibly customize NFV network services and finely control their behavior so as to expand the functions of NFV network services and improve their operation quality.The main work of this article includes:1.Aiming at solving the problems of inflexible control capabilities and inadequate scheduling behavior in SFC technology,the concept of Enhanced Service Function Chaining(ESFC)was first defined,and a flexible control of NFV network services according to user needs was proposed model;Secondly,it thesis the deployment network service and the description method of network service behavior,which is used to describe NFV network service requirements.Finally,a controller that converts the description file into machine-recognizable tasks is designed.2.A toolbox for setting up NFV network services task is designed and implemented.Based on the toolbox and controller,a system capable of controlling the timing and operating parameters of NFV network services was developed.The experimental results of the prototype system verify the feasibility of the ESFC mechanism.3.A visualization method of customized NFV network services is proposed.A virtualized test platform is constructed,and the test platform uses visualization methods to customize network services,such as Distributed Denial of Service(DDoS)attack and defense and low-speed DDoS attacks with visualization methods.Experimental results show that this method can significantly improve the efficiency of custom NFV network services.