Several data source discovery methods take into account the semantic heterogeneity problems by using several Domain Ontologies (DOs). However, most of them impose a topology of mapping links between DOs. DOs and mapping links are available on Internet but with an arbitrary topology. In this paper, we propose a data source Discovery method Adapted to any Mapping links Topology (DAMT) and taking into account semantic problems. Peers using the same DO are grouped in a Virtual Organization (VO) and connected in a Distributed Hash Table (DHT). Lookups within a same VO consists in a classical search in a DHT. Regarding the inter-VO discovery process, we propose an addressing system, based on the existing mapping links between DOs, to interconnect VOs. Furthermore, we adopt a lazy maintenance in order to reduce the number of messages required to update the system due to the dynamicity of peers. The performance analysis of the proposed method shows good results for inter-VO lookup queries. Also, it confirms a significant maintenance cost reduction when peers join and leave the system.
翻译:暂无翻译