SKIP: an efficient search mechanism in unstructured P2P networks

摘要: Abstract Search efficiency and accuracy of resource are important considerations for search algorithm in peer-to-peer (P2P) network. Most algorithms use flooding among neighbor nodes to relevant resource. However, this usually causes great amount redundant messages, which results high costs low precision. In paper, we vector space model (VSM) relevance ranking construct overlay network, a novel mechanism with K-iteration preference (SKIP) based on semantic group P2P networks is proposed efficiently solve these problems. The key idea SKIP reorder the neighbors according scores utilize selection during process query. We analysis implement scheme reveal that provides overhead topology maintenance, can be effectively used searching verify it outperformance higher precision lower cost by comparing current semantic-based gnutella-like efficient system (GES).