Searching Technology Based On P2P Network Structure

With the rapid development and popularization of internet, there are more and more information resources on line,so it is difficult to search, locate and retrieve resources. The emergence of searching engines solves this problem.However, because traditional searching engines adopt centralized mode, some problems still exist, such as outdated links, server failure, and limited storing capacity ,which seriously affect performances of searching engines. As a new network computing model , P2P has characteristics of self-organization, distribution and scalability. P2P technology brings new energy to search engines, which is applied to. All kinds of searching algorithms are brought forward and applied to practice.This paper puts forword a P2P searching mechanism based on peer interests. By digging out the peer interests, and forming a network according to the relevancy that they display, peer interests that have high relevancy will be close in the network. When searching information in P2P network, information searching will be routed to peers of relevant interests, so that searching efficiency is promoted.The main contents of this paper are concluded as following:(1)Firstly, P2P technology is summarized, including the concepts, characteristics, applications and searching technology, which is background information of this article.(2)A method is brought forward, which uses K-High Frequency terms Vector to express the peer interests, which has advantages such as low calculational quantity, good retractility and easy calculation of interest comparability, applicable in P2P condition.(3)Described in detail the topology of P2P network based on the peer interest. Entering arithmetic and exiting arithmetic are brought forward on the base of this mode.(4)P2P searching rooting algorithm based on peer interest is introduced. According to the relevancy of peer interest, peers that have similar resources will be close in the network, so that orderly organization of peer interest is realized. For a given inquiry ,the answer set is in the corresponding interest domain, thus search efficiency is greatly improved. As the seaching goes, more is learnt about other peers in the interested domain, searching efficiency is improved.(5)The JXTA architecture of Sun company was analyzed and studied, including the JXTA'three-layer architecture, basic concept and the six protocols.At last, on the basis of JXTA architecture, network structure, as well as relative searching functions, is designed and realized, and its performance is tested with experiments.