Improving The P2P Resource Locating Methods Based On Kademlia

In reeent years, with the development of the networks, new technologies are emerging and new directions are finding. At the beginning of the internet HTTP,FTP and SMTP were used, but now HTTPS,WEB2.0 and cloud computing are coming to us; From the beginning of the client\server model to the P2P strategy. All those changes make we realize that how to harness the data in the network adequately and reasonably is the driving force for the network.P2P network (Peer to Peer) is a self-organizing network, it can provide good technical support for peer to peer computing, collaborative work, search engines and so on. The resource location algorithm is a key research and hot spots in P2P network, while the structured distributed hash table algorithm for its unique high-performance and scalability advantages is Studied by more people. In these structured distributed hash table algorithm, Kademlia protocol for its stability, efficiency and download speeds is welcome, a lot of file-sharing systems are based on the Kademlia protocol.In this paper, according to the P2P network based on Distributed Hash Table (DHT), Flooding will be combined with the DHT system, the physical proximity of nodes in the network will be used to reduce the latency and optimize routing before the formation of the logical topology. An improved FKademlia based on original model is constructed. P2Psim is used to simulate the Performance of the new protocol and the Results show that FKademlia inherits the advantages of DHT and Kademlia, and it is superior to the original Kademlia model in the routing choices, the rate of successfully finding, the average logical path length and other aspects.