Super-Peer P2P Network Model And Its Search Algorithm Based On Semantics And Trust

With rapid development of network,P2P technology has become world-wide hotspot of computer network research.In a P2P system,each peer is both a client and a server,and all data interchange can be finished among the peers.Compared to traditional Client/Server(C/S)model,P2P is off-center,scalability,high performance-price ratio,strong robustness, load balancing.The fact that the construction of P2P overlay network doesn't consider the variety of shared contents in each peer makes the latency and the cost of locating resources increase.At the same time,the characteristics of dynamics and anonymity of P2P can't ensure that all the peers will provide honest services and reliable resources.Detailed research works have been done on the above issues.This thesis introduces a super-peer P2P network model based on semantics and trust(STS_P2P).STS_P2P clusters the nodes that have similar semantics.STS_P2P uses a two-layer based architecture.The upper layer consists of super-peers that form a super-peer domain.The lower layer consists of common peers which have similar semantics and form a semantic domain.STS_P2P adopts VSM(Vector Space Model)to define document vector,peer vector and group vector.The peers which have similar documents are clustered into same peer group according to the similarities of peer vectors.Trusts,capacity and stabilization are taken into consideration when select super-peers.In addition,a two-layer search mechanism that search in node and network are developed to process the queries.The results of the experiments proved STS_P2P architecture and the algorithms proposed.Benefiting from using the semantic information in network topology building and trust in query routing,STS_P2P acquired higher recall and utilization rate of bandwidth under the same restrict,in comparison with Gnutella network.Finally,a prototype system which includes file sharing,resource searching and downloading functions has been designed.