Multi-dimensional queries have been well-studied for client-server systems. How-ever, because of the popularity of distributed systems ( e.g., peer-to-peer systems, mobile ad-hoc networks), multi-dimensional queries in distributed systems are increasingly important. This thesis reports research results from two impor-tant related problems. First, we study the support of complex multi-dimensional queries, in particular, range and nearest neighbor queries, in peer-to-peer sys-tems. We achieve the goal by proposing a special adaptation of the R*-tree and the related algorithms. Second, we propose a distributed caching technique for multi-dimensional queries in mobile environments. It builds an index from previ-ously fetched query results which can be used to answer future queries. Finally,...[ Read more ]

Multi-dimensional queries have been well-studied for client-server systems. How-ever, because of the popularity of distributed systems ( e.g., peer-to-peer systems, mobile ad-hoc networks), multi-dimensional queries in distributed systems are increasingly important. This thesis reports research results from two impor-tant related problems. First, we study the support of complex multi-dimensional queries, in particular, range and nearest neighbor queries, in peer-to-peer sys-tems. We achieve the goal by proposing a special adaptation of the R*-tree and the related algorithms. Second, we propose a distributed caching technique for multi-dimensional queries in mobile environments. It builds an index from previ-ously fetched query results which can be used to answer future queries. Finally, we show the effectiveness of the proposed techniques with extensive experiments on real data.