Abstract

The web is often seen as the world's largest database and XML is regarded to provide its data model. As XML data is naturally distributed across the web it should be considered as a distributed database and subject to distribution design. The main tasks of distribution design are fragmenting the underlying database schema and allocating the fragments to different sites. The aim of fragmentation and allocation is to improve the performance and to increase the availability of data. Our objective is to generalise distribution design techniques from traditional databases to XML data. In this paper, fragmentation operations for XML documents are developed. As with relational databases, these operations include horizontal and vertical fragmentation. A third operation named splitting is borrowed from object-oriented databases and adapted to XML. We discuss how these techniques can be applied to the case of XML. Finally, we present a heuristic procedure for horizontally fragmenting XML documents, which is targeted at minimising the expected query costs.

Citations