Research On Routing And Caching Algorithm In Content Center Network

With some new network applications' appearance,the traffic of Internet is taking explosive increase,and the scalability,mobility and security problem of Internet is becoming more serious.In order to solve these problems fundamentally,the Content-Center-Network completely different from the TCP/IP network architecture came into being.The route forwarding of CCN adopts the method of facing the source server.This original route forwarding method tends to ignore the cache resources of the nearest storage node on the non-path node,resulting in a longer transmission delay.By default,CCN caches content on all nodes of the transmission path.This flooding caching method will result in a large amount of content redundancy on the nodes on the link.Therefore,in order to improve the performance of content center network,a better and more efficient CCN route forwarding and caching algorithm needs to be designed.For the problem which cache resource on non-path node is in low utilization rate leading to low cache hit ratio,this paper designed a route forwarding algorithm based on Improved K-means clustering Routing.Firstly,the IKR algorithm clusters all nodes in the network according to node relevance.Then,a control node is selected for each cluster according to the edge node condition and the node association degree condition to cache information of each node in the cluster.When the interest packet is forwarded,the node firstly matches its own cache table and pending request table.In the case that the two are not matched successfully,the control node in the cluster is queried,and the shortest path for obtaining the content data is found,thereby saving network resources.The simulation results show that,compared to full forwarding algorithm,random forwarding algorithm,shortest path forwarding algorithm,the IKR algorithm can effectively improve the cache hit rate and reduce the source server load and request average latency.For the problem that network nodes have a large amount of cache redundancy leading to long transmission delay in classic cache method,this paper designed a cache algorithm based on Node cache status and content popularity.The NCSCP algorithm is based on the improved K-means clustering model.In the process of interest packet forwarding,the cache node is selected according to the cache status and content popularity of the nodes on the path.When the packet is returned,the content is cached on the cache node selected when the interest packet is forwarded.For each content data,only one cache node is selected within the cluster to reduce cache redundancy within the cluster.The simulation results show that,compared with the local caching algorithm,probabilistic caching algorithm(probability is 0.5),weighted probability caching algorithm,NCSCP algorithm can effectively improve the cache hit rate and reduce the average request delay.