软件定义下的CCN缓存算法研究

为了解决内容中心网络（CCN）中存在的缓存问题，在内容中心网络和软件定义网络（SDN）的融合架构下，充分利用控制层全局感知和集中控制的优势，提出了一种软件定义下的CCN缓存决策方法。考虑了中介中心度、内容流行度、节点边缘度和缓存节点存储容量的有限性，引入了控制节点和通信节点的概念，通过控制器缓存决策，使得请求内容在不同的节点进行合理缓存。同时，根据控制节点距离缓存节点和服务器的跳数，找到返回用户请求的最短路径，采用ndnSIM进行仿真实验。结果表明，相比ALWAYS，LCD两种传统缓存决策策略，该方案可以明显提高缓存命中率，降低路径延展率和缓存替换数，达到了提高CCN缓存效率的目的,有利于内容中心网络及未来网络的发展。

Research on CCN cache algorithm under software definition

In order to solve the problems of caching in the content center network (CCN), based on the fusion architecture of content center network and software defined network (SDN), a method of CCN caching decision under the software definition is proposed, which makes full use of the advantages of global perception and centralized control in the control layer. The scheme considers betweenness centrality, content popularity, edge nodes and the limited storage of cache nodes at the same time, and it proposes the concepts of control node and communication node, so that the request contents can be cached reasonably at different nodes according to the controller''s cache decision. Meanwhile, the shortest path to the user request is found based on the hops of the control node to the cache node and the server. Simulation experiments are carried out using ndnSIM, and the results show that this scheme can obviously improve the cache hit rate, reduce the path extension rate and cache replacement number compared with the traditional caching strategies of ALWAYS and LCD, and the scheme achieve the purpose of improving the efficiency of CCN caching, which is beneficial to the development of the content center network and the future network.