Range-capable Distributed Hash Tables

In this paper, we present a novel indexing data structure called RDHT (Range capable Distributed Hash Table) derived from skip lists and specifically designed for storing and retrieving geographic data from a structured P2P network overlay. We have developed RDHTs as backend for the DART search engine, whose goal is to efficiently answer complex queries based on semantics and geographical context of the information stored in a P2P network. Queries are “range enabled”, in the sense opposite of the exact matching. Range and semantic queries on location based resources make it possible to answer questions such as “Where is the nearest bookshop?”. RDHTs merge the robustness and scalability of distributed hash tables with the simplicity and self maintenance of skip lists, while providing efficient support for range queries and proximity queries.