Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > Computer network architecture and design

Research on Resources Locating Strategy of Peer to Peer Network

Author ChengLan
Tutor ZuoJin
School Huaqiao University
Course Applied Computer Technology
Keywords Peer-to-Peer Network Power-law Small-world Apperceive Location Bloom Filter Probabilistic Distribution Queue
CLC TP393.02
Type Master's thesis
Year 2012
Downloads 9
Quotes 0
Download Dissertation

P2P (Peer to Peer) is one of the most important technologies that make the Internet’s fastgrowth. This technology makes use of the computing and storage resources that have fallen intodisuse. Every node in the P2P network can both get resources from other nodes and provideresources for them. This technology makes the resource sharing among computers in the Internetvery convenient and flexible. But the problem is: Large-scale using of P2P applications make thenetwork’s resource seem to be exhausting, increase costs of ISP, lower the network’s servicequality etc. Make a better p2p searching algorithm is necessary in order to lower the costs of P2Psearching in the network. How to utilize the efficient searching algorithm that can lower themessage generating, enhance the hit rate and reduce the response delay to optimize the P2Psearch model is a hot topic of P2P research field.The P2P Network can be classified as structured and unstructured due to its topology. Thispaper illustrates the current existing P2P network, mostly the existing unstructured P2P searchingalgorithm, and proposed two improved version of the unstructured searching algorithm.Based on the power-law distribution and small-world character of p2p network, thesearching algorithm which combines the method of apperceiving location of nodes with thestrategy of preferential attachment and a broadcasting strategy of interest attenuation mechanismin rumor spreading is proposed. The location of node is apperceived by analyzing theperformance information of the network layer’s path, and the broadcasting range is expanded byusing the degree information of node, also the probability model is established through rumorspreading mechanism. The performance of searching is improved by the probability methodwhich can fast guide the search to targeted node.The strategy of searching resource is a research hotspot in unstructured peer to peer network.It is hard to optimize response time, query hit, and coverage rate for resource location ofunstructured P2P network simultaneously. This paper presents a quickly search algorithm calledBFPDQ (Bloom Filter and probabilistic distribution queue), which is based onprobabilistic distribute queue and Bloom Filter technology. BFPDQ is mainly used foracyclic random network. Information of resources and requests can be expressed by Bloom Filtertechnology. Meanwhile, performance information of the underlying network’s path can be usedto guide transmitting strategy for upper layers. PDQ (probabilistic distribute queue) usesdistributed queues to substitute traditional walkers to search resources. Requester coordinatesdirection and depth of those queues and aggregates their resource location messages. This paperimproved the search forwarding mechanism, more conducive to hit the target node.According to the above analysis, this article improved the feasibility of search strategy andsimulated experiments are carried out to verify.

Related Dissertations
More Dissertations