On the Topology Maintenance of Dynamic P2P Overlays through Self-Healing Local Interactions

Abstract

This paper deals with the use of self-organizing protocols to improve the reliability of dynamic Peer-to-Peer (P2P) overlay networks. We present two approaches, that employ local knowledge of the 2nd neighborhood of nodes. The first scheme is a simple protocol requiring interactions among nodes and their direct neighbors. The second scheme extends this approach by resorting to the Edge Clustering Coefficient (ECC), a local measure that allows to identify those edges that connect different clusters in an overlay. A simulation assessment is presented, which evaluates these protocols over uniform networks, clustered networks and scale-free networks. Different failure modes are considered. Results demonstrate the viability of the proposal.