Research On Service Function Chain Deployment Optimized For Security

Network function virtualization technology has brought a lot of convenience to our lives,and at the same time security issues have gradually surfaced.Security plays an important role in our daily life,so how to optimize the security performance of services while ensuring resource allocation and other issues is of great significance.This thesis redefines the security parameters.The security parameters are no longer static,but are based on the deployment time and the diversity of VNFs deployed on the physical server,and the security performance is quantified according to this parameter.With the advent of services,the security parameters of the underlying network will inevitably decrease.How to balance the average security performance of the underlying physical network is also a problem worthy of further consideration.The rapid development of modern networks makes the underlying network no longer only a single domain.Multi-domain networks are also common in reality.How to optimize security performance under single domain and multi-domain is also of great significance to operators.Therefore,this paper mainly studies security performance optimization and solves the problem of service function chain deployment in single-domain and multi-domain situations.The main contributions and innovations are as follows:(1)Introduce dynamically changing security parameters to a single domain,and design deployment algorithms and migration algorithms to optimize security performance.The algorithm mainly includes the calculation and update of security parameters.On the premise of optimizing the security performance,the deployment algorithm is designed to deploy SFC,and the migration algorithm is designed to improve the average security performance of the underlying network.After simulation tests,the algorithm has significantly improved the deployment success rate,average initial deployment security parameters,average security parameters,and resource utilization compared to the comparative algorithm.It improves the deployment success rate and resource utilization while achieving optimized security performance.aims.(2)Introduce dynamically changing security parameters to a multi-domain network,and design cut-chain and sub-chain deployment algorithms and migration algorithms based on the characteristics of the multi-domain topology to improve security performance.The algorithm mainly includes the topology abstraction method based on topology aggregation technology proposed for the issue of inter-domain privacy,and the cut-chain method based on the average security parameter ratio and the sub-chain deployment algorithm to optimize security performance.Finally,a migration algorithm is designed to improve average security parameter.After simulation tests,the algorithm not only achieves the goal of optimizing security performance,but also has better performance than the comparative algorithm,while reducing the algorithm response time.At the same time,it also improves the deployment success rate and deployment overhead compared to the comparative algorithm.