Building self-adaptive Peer-to-Peer overlay networks with dynamic cluster structure

The tremendous growth of public interest in Peer-to-Peer (P2P) networks in recent years has initiated a lot of research work on how to design efficient overlays for P2P systems. Among them are some overlays that support range queries, but their basic topologies might be different with diverse options, such as ring, cluster, binary tree and doubly-linked lists, etc. In this paper, we present a self-adaptive P2P overlay with dynamic cluster structure that supports range queries. In such cluster-based topologies, a challenge issue is how to choose appropriate cluster size in order to ensure that the least total messages are produced in the network. Inspired by evolutionary game theory, this paper presents a novel approach for improving query routing performance in a fully distributed and self-adaptive way where super-peers can modify their behaviors (merging or splitting clusters) with incomplete information and bounded rationality for the purpose of improving their individual benefit. Extensive experiments have been carried out to assess the performance of the proposed technique. The obtained results have highlighted remarkable benefits under churn in terms of bandwidth consumption.