Research On The Consistency And Query In P2P System

Peer-to-Peer system is an Internet application that can take full advantageof resources of autonomous participant. With early successful paradigms likeNapster and Gnutella, research communities have dedicated themselves todiscover the substantial characters and actualized techniques. For the goals ofpresenting a tool for mapping table's inference and consistency and a solutionfor translation a local query. The paper does the in-depth research on somepivotal issues about data sharing.Firstly, the paper introduces the importance of inference and consistencyand presents a partition algorithm to solve the inference and consistencyproblems effectively, and later discusses the problems of executed conditionsand complexity.Secondly, the paper focuses on the issue of the query translation. By meansof adding the module of SQL parser into the original architecture, the outerqueries are converted to the inner tableau information to be stored later in thecomputed mapping tables as the outputs of above partition algorithm. Thepaper presents an algorithm of complete translation for local query andvalidates its soundness theoretically.Finally, exhaustive experiments are designed to evaluate the executivecharacter, performances with respect to several parameters, of which the testingdata are retrieved from the European biological databases. At the end of thepaper, it analyzes the experimental results and draws a conclusion, all of whichprovide the future work with the abundant experiences and a foundation.