During the last decades, the vast number of independent data sources and the high rate of data generation make central assembly of data at one location infeasible. As a consequence, data management and storage become increasingly distributed. Although the client-server architecture as communication model is still popular for application development, traditional client-server solutions are prone to bottleneck risks and therefore do not scale as the number of participating sources increases. Peer-to-peer (P2P) systems comprise a scalable collaborative architecture of autonomous participants (called peers), where each peer serves both as a client and as a server. As data storage becomes inherently distributed, an emerging challenge is to support efficient query processing over data stored at disparate locations, which allows users to discover and retrieve relevant data. Even more important is to incorporate and support flexible query operators, such as similarity search, skyline and top-k queries, that help avoiding huge and overwhelming result sets. Moreover, as data in modern applications is typically multidimensional, effective distributed query processing methods and techniques that apply to multidimensional data are sought. Query processing in P2P networks poses inherent challenges and demands non-traditional techniques due to the distributed nature of the environment and lack of global knowledge. This chapter provides an introduction to the main topics and basic concepts related to P2P query processing over multidimensional data.
Unable to display preview. Download preview PDF.