Research On Technology Of Cooperative Caching Based On Centralized Control

With the rapid development of network technology and the explosion growth of information, the Internet has undergone tremendous changes. Network size is growing. More and more rich business needs.The number of users is more and more great. The traditional host to host network architecture has been difficult to meet the development of the future network. The Information-Centric Networking and Software Defined Networking come into being. The Information-Centric Networking is a network architecture that is based on information rather than host. Everything is information in Information-Centric Networking.The network connect through information. The core object of Information-Centric Networking is information. The Software Defined Networking is a network architecture which separates control plane from data plane and has programmable features.We propose a centralized control caching strategy based on popularity and betweenness centrality (CORPORATE). In CORPORATE,different from traditional caching strategies, we define a control node on the basis of the betweenness centrality to cache the content. Due to the tremendous amount of content in the internet and the limit of the caching node storage, the content cached at the control node is replaced frequently.Then the content is moved into the surrounding node based on popularity and betweenness centrality. Meanwhile, according to the hops of the control node to the caching node and the server, the control node can get the shortest path between the consumer and the content providers. We loop the above process gradually moving the content to the edge node,which can greatly improve the average link load and the average transmission delay. Based on the centralized control characteristic of the Software Defined Networking, the nodes are connected with each other and the cache is moved to the edge user gradually. Based on Icarus simulation platform, the simulation system environment is set up and the performance of CORPORATE algorithm is analyzed and evaluated. the simulation results demonstrate that COPRORATE can greatly improve the average link load and the average transmission delay.