Research On Deployment Of Virtual Network Function For Resource Optimization In SDN

With the rapid development of network technology,the number of Internet users and users' demand for business traffic have increased dramatically.Therefore,network operators need to continue to expand the construction of network infrastructure to meet users' demands for business traffic,which makes the network scale continue to expand and causes many problems.Software defined network(SDN)technology and network function virtualization(NFV)technology decouple network functions from dedicated hardware devices and deploy them to commercial servers in the form of software.In the NFV network,user service requests usually need to pass through a set of virtual network functions(VNF)in a specific order,such a set of VNFs form a service function chain(SFC).However,the resources in the network are often limited,and it makes sense to implement resourceoptimized virtual network function deployment.This thesis mainly studies the deployment of virtual network functions of service function chains in networks that introduce NFV technology.The main research contents are as follows:1.Aiming at the current deployment method considering the single problem of network resource utilization,this thesis considers the utilization of link and node resources to achieve the purpose of optimizing the use of the entire network resource,and designs a heuristic search service function chain deployment method.First,the network model of service function chain deployment is described,and the mathematical model of service function chain deployment is established.Then,the link bandwidth resources and node computing resources are respectively portrayed in the form of relative cost of resources,and the relative cost of link and node resources is minimized as an optimization goal.A heuristic search service function chain deployment algorithm is proposed to achieve optimized deployment of service function chain,the algorithm consists of two stages of virtual network function deployment and link mapping.The results of simulation experiments show that the method in this thesis can fully utilize the network resources while meeting the service needs of users,and improve the request acceptance rate and resource utilization rate.2.Aiming at the problem of service function chain deployment,another deployment method for optimizing resources is designed.This method takes computing resources and link resources as a whole,minimizes the relative cost of the two resources as the optimization goal,and performs mathematical modeling.According to the Hidden Markov characteristics of the VNF deployment problem in the service function chain,a service function chain deployment of resource optimization method based on Viterbi algorithm is proposed.This method selects the path with the smallest total cost of resources that meets the constraints based on link resources and node resources,and obtains the deployment plan by backtracking the nodes in the shortest path.The results of simulation experiments show that this method is not only better than the other deployment methods compared,but also slightly better than the method proposed in the thesis,and its request acceptance rate and resource utilization rate are further improved.