Grouped Peer-to-Peer networks and self-organization algorithm

Locating contents is an essential function, but a very difficult and challenging problem for large-scale Peer-to-Peer (P2P) systems. For this problem, many P2P systems, architectures, and search strategies are proposed. In this paper, we focus on the self-organization of a group structure based on the local interest for P2P systems and propose a self-organization algorithm. We evaluate the performance of the self-organized network through simulations. These results show that 1) the query hit rate is improved and 2) networks have robustness for the change of size of local storage shared by each peer.