Targeted Local Immunization in Scale-Free Peer-to-Peer Networks

The power-law node degree distributions of peer-to-peer overlay networksmake them extremely robust to random failures whereas highly vulnerableunder intentional targeted attacks. To enhance attack survivability ofthese networks, DeepCure, a novel heuristic immunizationstrategy, is proposed to conduct decentralized but targetedimmunization. Different from existing strategies, DeepCure identifiesimmunization targets as not only the highly-connected nodes but also thenodes with high \it availability and/or high \it link load, with theaim of injecting immunization information into just \it right targetsto cure. To better trade off the cost and the efficiency, DeepCuredeliberately select these targets from 2-\it local neighborhood, aswell as topologically-remote but semantically-close friends if needed.To remedy the weakness of existing strategies in case of sudden epidemicoutbreak, DeepCure is also coupled with a local-hub oriented \it ratethrottling mechanism to enforce proactive rate control. Extensivesimulation results show that DeepCure outperforms its competitors,producing an arresting increase of the network attack tolerance, at alower price of eliminating viruses or malicious attacks.