Performance evaluation of gossip protocols on complex networks.

This paper presents a study on gossip protocols employed to spread data in unstructured Peer-to-Peer (P2P) network overlays. We consider two classic protocols (fixed probability and probabilistic broadcast) and compare them with two new degree dependent gossip protocols that change the probability of dissemination based on the degree of peers. We identify principal metrics, provide a theoretical model and perform the assessment evaluation using parallel and distributed simulation. A main point of this study is that our large-scale simulations consider implementation technical details, such as the use of caching and Time To Live (TTL) in message dissemination, that are usually neglected in simulation, due to the additional overhead they cause. Outcomes confirm that these technical strategies have an important influence on the performance of gossip schemes. Moreover, gossip schemes are quite effective to spread information in P2P overlay networks, whatever their topology.