Research On The Push Mechanism And Cache Consistency Based On Hyperbolic Routing For Ndn

Named Data Networking(NDN)is the most representative member of the information center network.NDN has changed the transmission mode of traditional IP network.It realizes distributed information exchange by taking advantage of the feature that routing nodes can store data.NDN naturally supports multicast and multi-path transmission mode,and its intermediate cache can greatly reduce delay.NDN is now widely used in the Internet of thing and sensor network.However,its inherent architecture only supports Pull communication mode,and distributed storage also brings the problem of cache inconsistency.In recent years,Push communication becomes more and more important,and the problem of cache consistency needs to be solved urgently.This paper proposes a Push mechanism based on hyperbolic routing.And combined with the Push mechanism and Zipf distribution,this paper proposes a cache update strategy to achieve provider-driven.The HyperPush mechanism designed in the first part of this paper is different from the current Push mechanism,which utilizes the characteristics of greedy addressing of hyperbolic routing to solve the problem that producers cannot find consumers in NDN.This paper proposes a hyperbolic embedding optimization algorithm,called Pop-Hyper,which takes popularity of content as the scale of coordinate distribution to make it more corresponds to NDN.HyperPush changes broadcast flooding push into a purposeful multicast push,which greatly reduces the network load.In this article,a new Pending Data Table(PDT)structure is added to record the data list to be pushed,and the Interest structure is modified to distinguish between the two communication modes of Interest packages.The simulation experiment with ndnSIM shows that HyperPush can greatly reduce the communication delay,increase the cache hit rate and reduce the overload.In the second part of this paper,the POPcache update strategy is proposed.According to Zipf's law distribution,this paper achieves strong consistency with high popularity.Meanwhile,based on network parameters such as overload,delay,cache capacity and popularity of content,a mathematical model is established to determine the optimal cache location of data,so as to reduce the load caused by updating.The simulation results of ndnSIM show that the POPcache update strategy eliminates the reliance on freshness value and improves the content consistency within the tolerable range of load increase.