The Study Of Searching Algorithm Of The P2P Networks

In recent years along with the emergence of many P2P systems, the P2P technique becomes the researchful hotspot. The P2P technique is based on the equal thought and this thought is changing people's thinking about the network model. Someone forecasts the P2P network model will become the new network model of next generation. Now the P2P technique is mainly applied in file sharing, distributed calculation, cooperating systems, electronic commerce and search engine. The P2P technique will be used extensively along with the research.The file sharing is the reason the P2P technique springs up and is also the broadest application of the P2P technique. The file sharing provides convenience for the large of End-User. Everyone can share his files to other people and also can download files from other people. The p2p systems become popular because of the application. So it did not consider the performance problems in design, such as the security, the scalability and so on. And now many people carry on an investigation in the P2P systems' performance.Gnutella is the second generation p2p system. It is a simple system and is widely applied at present. Gnutella produces large numbers of messages in the system when locate a file. So there is a great deal of bandwidth consuming and the system is not scalable. In this paper, we study the searching mechanism of Gnutella and the related research. Based on the study, we present an intelligent searching algorithm. Our algorithm adjusts the topology of the network base on the bypast query messages. So the number of the messages is reduced when we locate a file in the system. The response time is shortened and the scalability is improved. At the same time, we use the active flow control to restrict thenumber of the connections of each peer and avoid the peers overloading.