A Distributed Hybrid Scheme for Unstructured Peer-to-Peer Networks

Peer-to-peer (P2P) networks have gained a lot of popularity in recent years. While structured (DHT) networks provide better response time and smaller diameter, which are the advantages over unstructured networks, they are vulnerable to frequent node failure/joins/leaves. There is always a tradeoff for any P2P network to achieve all of these goals. In this paper, a distributed hybrid scheme is proposed for unstructured P2P networks, which combines Markov random walks and peer clustering to achieve a better tradeoff. The scheme has a short response time for most of the queries that belong to the same interest group, while still maintaining a smaller network diameter. More important, we propose a totally distributed clustering algorithm, which means better resilience to network dynamics. The performance of our systems is demonstrated by extensive simulations.