Research On Cache Management Scheme In ICN

Information-centric networking(ICN),as a completely new communication network model,it has become a hot spot in research of future network architecture in recent years.ICN is information-centric,uses name of information to identify,retrieve,and route the data.ICN uses the cache as a built-in structure.By default,all routers potentially store all the data passing through,so that subsequent responses can be responded as quickly as possible.As a basic component of ICN,the caching system plays a vital role in the content distribution service.Its efficiency directly affects the performance of the entire network.How to uniformly schedule and manage cache resources in the network and design efficient cache management methods is an important topic in the research of ICN.How to make use of the limited cache resources of network,and improve cache resource utilization,that is the main purpose of research on cache management method.Existing edge caching mechanisms deploy content at network edge as much as possible and make the content close to users in order to reduce the user's request-response delay and improve the users' experience,but weaken reusability of cached content and reduce the probability of cache responding to requests from nodes in other locations of network.In the core caching methods,copies of content is deployed at the core nodes of network and can respond to requests from nodes in different regions,but they are far away from users.The response delay is large.How to design a cache management method to improve performance of cache system in ICN,this article conducts the following research.For the problems of edge cache,a cache decision mechanism based on the request popularity of information is proposed.The cache hit ratio is improved by increasing reusability of cache contents,that is,the response probability of cached contents for same request on other paths is increased.The frequent access of nodes to specific content in the network is time sensitive.For this feature,a cache replacement mechanism based on the popularity of access is proposed.Based on the popularity of content access by nodes,it perceives changes in the popularity of content in time and adjust cache deployment.Simulation experiments have verified that this mechanism can reduce cache redundancy,improve cache hit rate without increasing response delay.There may be some correlation between the contents accessed by nodes.Taking advantage of this feature,a path caching method based on content correlation is designed.By discovering the correlation between target content and the popular contents of the node,and considering the position of the node on path,to determine whether the current node caching received data.The feasibility and effectiveness of the method are verified by simulation under different scenes.