XML Multidimensional Modelling and Querying

TL;DR

This paper investigates XML-based multidimensional data modeling and OLAP query processing, comparing approaches, demonstrating query expression techniques, and evaluating performance improvements with extended XQuery features.

Contribution

It introduces methods for modeling multidimensional data in XML, enhances OLAP query expressiveness with XQuery 1.1, and evaluates performance across different models using an extended database.

Findings

XQuery 1.1 grouping features improve query performance

Different XML multidimensional models vary in query efficiency

Extended eXist database supports advanced grouping in XQuery

Abstract

As XML becomes ubiquitous and XML storage and processing becomes more efficient, the range of use cases for these technologies widens daily. One promising area is the integration of XML and data warehouses, where an XML-native database stores multidimensional data and processes OLAP queries written in the XQuery interrogation language. This paper explores issues arising in the implementation of such a data warehouse. We first compare approaches for multidimensional data modelling in XML, then describe how typical OLAP queries on these models can be expressed in XQuery. We then show how, regardless of the model, the grouping features of XQuery 1.1 improve performance and readability of these queries. Finally, we evaluate the performance of query evaluation in each modelling choice using the eXist database, which we extended with a grouping clause implementation.