Duty-oriented hybrid peer-to-peer network based on metadata

Abstract

The topology structure of the peer-to-peer overlay network is a critical factor in dominating performance of resource discovery. In the paper, we consider that nodes have their duty or function in the real world, at the same time, we take full advantage of the characteristic of unstructured peer-to-peer network. According to these, a duty-oriented hybrid peer-to-peer network based on metadata(DHPM) is proposed. In the tree-like network structure, A 1st level super peer existing in one domain is determined according to duty or function, while 2nd level super peers existing in groups and peers are decided by the semantic similarity. Metadata directory mechanism is used to maintenance the overlay network and to query the resources. Node join algorithm, node departure algorithm and query searching algorithm are presented to ensure construction and running of the network. Finally, experiments are made to show the performance of query searching with DHPM, Gnutella and Freenet.