OLAP Presentation Modeling with UML and XML

On-Line Analytical Processing (OLAP) is a trend in database technology, based on the multidimensional view of data. Multidimensional data (called multicubes) form the basic logical data model for OLAP applications and a significant number of proposals exist for cube models, without – in most cases – the distinct description of an equivalent “presentational” layer or model for the presentation of multidimensional data and their hierarchies. In this paper we show how a formal presentation model for multidimensional data, namely the Cube Presentation Model (CPM) can be modeled using an extension of UML, by means of stereotypes, to more elegantly represent the most common presentational aspects of OLAP. Further more, we exploit the eXtensible Markup Language (XML) to capture the necessary semantics of the presentation model that can be used in OLAP visualization applications. Finally, we demonstrate all these using Rational Rose.