论文部分内容阅读
In order to improve the performance of peer-to-peer files sharing system under mobile distributed en-vironments,a novel always-optimally-coordinated(AOC)criterion and corresponding candidate selectionalgorithm are proposed in this paper.Compared with the traditional min-hops criterion,the new approachintroduces a fuzzy knowledge combination theory to investigate several important factors that influence filestransfer success rate and efficiency .Whereas the min-hops based protocols only ask the nearest candidatepeer for desired files,the selection algorithm based on AOC comprehensively considers users’ preferencesand network requirements with flexible balancing rules.Furthermore,its advantage also expresses in theindependence of specified resource discovering protocols,allowing for scalability.The simulation resultsshow that when using the AOC based peer selection algorithm,system performance is much better thanthe min-hops scheme,with files successful transfer rate improved more than 50% and transfer time re-duced at least 20%.
In order to improve the performance of peer-to-peer files sharing system under mobile distributed en-vironments, a novel always-optimally-coordinated (AOC) criterion and corresponding candidate selection algorithms are proposed in this paper. Compared with the traditional min-hops criterion, the new approachintroduces a fuzzy knowledge combination theory to investigate several important factors that influence filestransfer success rate and efficiency .Whereas the min-hops based protocols only ask the nearest candidatepeer for desired files, the selection algorithm based on AOC comprehensively considers users’ preferencesand network requirements with flexible balancing rules. Futuremore, its advantage also expresses in the independence of specified resource discovering protocols, allowing for scalability. The simulation resultsshow that when using the AOC based peer selection algorithm, system performance is much better thanthe min-hops scheme, with files successful transfer rate improved more than 5 0% and transfer time re-duced at least 20%.