Adaptive search for peer-to-peer networks

Nowadays, the most popular peer-to-peer applications operate on unstructured networks. The obvious problems that have infected the popularity of peer-to-peer (P2P) systems are effective information search and traffic caused by the blind flooding-based search. In this paper, we have concentrated on developing an agent-based model for controlling query messages that are represented as agent; the ecologically inspired distributed search (EIDS), which derives its inspiration from natural ecosystem, is presented. We models autonomous adaptive agents after several ecological concepts and mechanisms. As an experiment result, we compare its performance against the well-known k-random walker approach. We focus on the problem of actively changing the topology of the P2P network by utilizing Schellingpsilas segregation model to improve the efficiency of search.