Research On Neighbor Object Availability-based Cooperative Caching Policy For ICN

Internet usage has drastically shifted from host-centric end-to-end communication to receiver-driven content retrieval.In order to adapt to this change,a handful of innovative Information-centric Networking(ICN)architectures have recently been proposed.One common and important feature of ICN architecture is to leverage built-in network caches to improve the transmission efficiency of content dissemination.However,the default caching placement policy introduces high redundancy,and the caching resources cannot be utilized fully,and the default routing lookup mechanism does not utilize the transient cache content of the nodes in the network,and cannot reduce the network latency drastically.Aiming at the above problems,this dissertation optimizes the caching placement strategy and routing lookup mechanism of the Information-centric Networking,and particular focus on reducing cache redundancy and improving the availability of cached content.The main work includes the following:(1)The default cache placement policy introduces high redundancy,and the cache resource utilization is low.Besides the dynamic of the network is not considered.Aiming at the above problems,this dissertation proposes a ICN cooperative cache placement policy based on node hotness and cache replacement rate,which further reduces the number of request hops and improves the benefits of the caching system.(2)The default cache placement policy does not take advantage of cache resources on the neighbor nodes,and the appropriate placement policy is not used for content of different popularity.In order to solve the above problems,this dissertation proposes a ICN neighbor collaborative caching placement strategy based on content popularity and node hotness,and improves the utilization of cache resources.(3)The default routing lookup mechanism does not utilize the transient cache resources of the neighbor nodes,only establishes the routing entries for stable content sources.Against the above problems,this dissertation proposes a ICN neighborhood routing strategy based on the Bloom filter,which uses Bloom filter to store the neighbor node cache content information while utilizing the less cache space.The simulation results show that the policies proposed by this dissertation can obtain lower content request latency,lower network traffic,and lower server load.The main innovations of this dissertation includes the following:(1)The concept of node hotness is used according to the dynamic nature of the network;(2)The statistic model of the content popularity is proposed according to the characteristics of the low-pass filter with the node;(3)Use the Bloom filter to store the information of the neighbor node cached content,occupy less cache space while speeding up content lookup speed.