HAND: an overlay optimization algorithm in peer-to-peer systems

Recently, peer-to-peer (P2P) model becomes popular due to its outstanding resource sharing ability in large-scale distributed systems. However, the most popular P2P system – unstructured P2P system, suffers greatly from the mismatching between the logical overlay and physical topology. This mismatching problem causes a large volume of redundant traffic, which makes peer-to-peer system far from scalable and degrades their search efficiency dramatically. In this paper we address the mismatching problem in an unstructured P2P architecture by developing a distributed overlay optimizing algorithm – HAND (Hops Adaptive Neighbor Discovery). Through comprehensive simulations, we show that HAND significantly outperforms previous approaches, especially in large-scale P2P systems.