ACO-Based Search Algorithm in Unstructured P2P Network

Flooding-based search mechanism has been used extensively in unstructured peer-to-peer network such as Gnutella, but the mechanism is not scalable and as a consequence, it consumes a high amount bandwidths and resources. In order to solve the problems, an ACO-based search algorithm is used in this paper. By introducing Ant Colony Optimization, each peer maintains routing table, which store the amount of pheromone corresponding to classification dropped at the link. Based on the pheromone values, a query is flooded to those peers which are most likely to be resources owner. The update of phenomenon depends on the number of documents found and the link cost in query by all ants´ collective cooperation. Simulation results show that, compared with Modified-BFS mechanism, the ACO-based search algorithm can effectively improve the search performance, and becomes better than the Modified-BFS as the peers optimize their routing tables while using a much smaller number of messages.