Font Size: a A A

Research On Fault Tolerance For Hybrid P2P System Based On Backup Super Peers

Posted on:2010-02-19Degree:MasterType:Thesis
Country:ChinaCandidate:Y S HuFull Text:PDF
GTID:2178360275473709Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
In recent years, Peer-to-Peer technology has been widely used, and becomes the main application types that consume a large fraction of Internet traffic. With the rapid development of P2P technology, it has been concerned by computer industry. How to improve the rate of successful query, and enhance the capacity of fault tolerance, has become the main work of P2P network. This paper will research the fault tolerance of Super-peers P2P network which is a kind of Hybrid P2P network.In this paper, we analyzed the fault tolerance capability and the technology of fault tolerance of three types of P2P network topology; and then compared the technology of fault tolerance of hybrid P2P network from fault tolerance capability and fault tolerance overhead; improved a fault tolerance mechanism based on backup super peers: Start from the mind of Reputation-Aware Algorithm, select a part of peers from the group to be the backup super peers; make sure all the backup super peers are in a bi-connected state; add monitoring mechanism to the backup super peers to monitor the state of super peer. If the super peer fails one of the backup super peers will receive the message. This peer will use breadth-ergodic in backup super peer bi-connected graph to choose a backup super peer with the highest reputation. The chosen peer will become a new super peer, to provide services for peers in the group.In this paper, we have carried out a simulation and a test of the above fault-tolerant mechanism. The result shows that after the application of the fault-tolerant mechanism, the query success rate of P2P system is greatly improved, while the fault tolerance overhead is not too much.
Keywords/Search Tags:Hybrid P2P Network, Backup Super-Peer, Fault-tolerance Capability, Bi-connected Graph, Monitoring Mechanism
PDF Full Text Request
Related items