Performance Research Of Performance Evaluation Of Content Distribution Network Based On PEPA Model

With the rapid development of information technology,the Internet has become one of the major infrastructures of the information society.But after more than half a century of development,the Internet has evolved into a complex giant system.The contradiction between limited cyber source and growing user demand have become increasingly prominent with the rapid growth of users resources’ size and significant concurrent access.At the same time,with the arising of a variety of new applications and intelligent terminals,when various business forms access heterogeneous resources,requirements of service quality vary greatly.The traditional content delivery method is difficult to meet the requirements,which makes the content distribution face major challenges.Content distribution network(CDN),which was born in 1998,can ease the bottlenecks of Internet content provider through the copy of the content and users access in the nearest way.After many years of exploration and practice,CDN technology has received extensive attention and got rapid development,in which its distribution service has become an important part of the Internet.CDN technology can provide network infrastructure and operation mechanism in a scalable way to efficiently distribute content and services so as to enhance the users’ network experience.It has been shown that CDN technology can effectively deal with the challenges of content distribution by improving the structure of CDN,optimize the performance.Firstly,this paper proposes a stochastic process algebra method named performance evaluation process algebra(PEPA),which is suitable for modeling and analysis of the system,and specialized evaluate the performance of the system.In this paper,for each component in the content distribution network and the cloud based content distribution network,the corresponding PEPA model are established,which can better describe the data transmission process between the components in the content distribution network,so that the workflow in the network model can be analyzed.In this paper,we simulate the response time of user access in content distribution network and cloud content distribution network.At the same time,a special algorithm is also given to simulate the response time,whose core idea is that in the premise of the special action flow,it sets the mark to indicate the completion of the action to Makes the marked action correspond to the action stream until the end of a complete action path;meanwhile,the fluid approximation method is introduced In this paper.By this method,this paper gives the ordinary differential equation(ODEs)of the related components in the content distribution network and the cloud content distribution network.The throughput and utilization of the related components in the network can be calculated by these equations,and through the analysis and evaluation of response time,throughput and utilization ratio,the content distribution network framework model can be improved accordingly in the future.Secondly,an improved algorithm for simulating response time in content distribution networks and cloud based content distribution networks is proposed In this paper.Due to the limitation of these special algorithms,a general algorithm is presented to simulate the response time,in which the corresponding response time in the case of any action flow can be simulated,and a model case is used to simulate the response time by this algorithm.Finally,the Little theorem is verified in PEPA model in this paper.The validation of the method is based on a combination of prior knowledge of the Little theorem in the closed system and the relevant knowledge of the PEPA.This theorem can provide a theoretical basis for solving the average response time of PEPA modeling system.