基于谣言传播机制的P2P搜索算法研究-计算机应用技术专业论文.docxVIP

西南交通大学硕士研究生学位论文 西南交通大学硕士研究生学位论文 第ll l页 Abstract Allowing file exchange among the end client peers through Intemet is the initial purpose of P2P(Peer-to-Peer)．In the short history,P2P application has become one of the main application types that consume a large fraction of Internet扛攒c．硼1c capabi．ty that P2P architecture supports massive BSerS makes it suitable for rapidly deploying powerful and large—scale distributed applications with low cost． In order to take full advantage of a variety of r懿OIlrc铝in P2P networks,the most important thing iS to find those resourc铝effectively． However,because P2P network is dynamic and scalable，it iS dimcult to design an excellent search algoritltm如r P2P．Rumor or Gossip algorithms are a class of algorithms for message muting．They are also called epidemioMgieal algorithms，since messages are spread in a network much like a disease in a susceptible population,that is，the neighbo硌to which messages ale forwarded by each node age cboSan at random．Comparing with the flooding algorithms，the rumor algorithms Can decrease the cost in large quantity．But because of the randomness of the message spread,the algorithms can not assllre the coverage ratio of spread．and mcrease tlle time that message propagates． In order to overcome the shortage of rumor algorithms，the author improved the traditional Rumor algorithms by utilizing the Gnutellas Power-law characteristics．Instead of selecting neighbor at random，the improved algorithms propagate message based on the degree ofnode，which overcomes the limit of selecting neighbor at random，and thus optimizes the performance of Rumor algorithms．The feasibility and affeetivity of the improved algorithm arc proved by using Gnutellasim,a type of the simulation sothⅣare for Gnutella networks，to sknulate the improved and the original algorithms． The main research work is as follows： (1)Introducing some basic notions and knowledge of P2P networks． 西南交通大学硕士研究生学位论文 西南交通大学硕士研究生学位论文 第1 V页 including the definition an