2006 ResearchinDataWarehouseModeling

(Rizzi et al., 2006) ⇒ Stefano Rizzi, Alberto Abelló, Jens Lechtenbörger, and Juan Trujillo. (2006). “Research in Data Warehouse Modeling and Design: Dead Or Alive?.” In: Proceedings of the 9th ACM international workshop on Data warehousing and OLAP. doi:10.1145/1183512.1183515

Subject Headings: Metadata Modeling, Data Warehouse Modeling.

Notes

Cited By

Quotes

Author Keywords

data warehouse design; multidimensional modeling

Abstract

Multidimensional modeling requires specialized design techniques. Though a lot has been written about how a data warehouse should be designed, there is no consensus on a design method yet. This paper follows from a wide discussion that took place in Dagstuhl, during the Perspectives Workshop "Data Warehousing at the Crossroads", and is aimed at outlining some open issues in modeling and design of data warehouses. More precisely, issues regarding conceptual models, logical models, methods for design, interoperability, and design for new architectures and applications are considered.

…

5. Interoperability and Metadata

The heterogeneity in conceptual and logical models proposed for DWs, together with the wide variety of tools and software products available on the market, has lead to a broad diversity in metadata modeling. In practice, tools with dissimilar metadata are integrated by building complex metadata bridges, but some information is lost when translating from one form of metadata to another. Thus, there is a need for a standard definition of metadata in order to better support DW interoperability and integration, which is particularly relevant in the recurrent case of mergers and acquisitions.

Two industry standards developed by multi-vendor organizations have arisen in this context: the Open Information Model (OIM) [52] by the Meta Data Coalition (MDC) and the Common Warehouse Metamodel (CWM) [56] by the OMG (see [80] for a comparison of the two competing specifications). In 2000, MDC joined OMG for developing the CWM as a standard metadata model. The CWM is a platform-independent metamodel definition for interchanging DW-specifications between different platforms and tools. It is based on the standards UML, XMI, and MOF, and basically provides a set of metamodels that are comprehensive enough to model an entire DW including data sources, ETL, multidimensional cubes, relational implementations, and so on. These metamodels are meant to be generic, external representations of shared metadata and to provide a framework for data exchange. Unfortunately, their expressivity is not sufficient to capture all the complex semantics represented by conceptual models, so they hardly can be used for effective integration of different DWs.

An alternative approach in this direction is described in [8], where a notion of dimension compatibility based on information consistency is proposed, aimed at cross-querying over autonomous, federated data marts. We believe that another interesting possibility for integration would be to use domain ontologies in order to establish semantic mappings between different data marts.

References

,

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2006 ResearchinDataWarehouseModeling	Stefano Rizzi Alberto Abelló Jens Lechtenbörger Juan Trujillo			Research in Data Warehouse Modeling and Design: Dead Or Alive?				10.1145/1183512.1183515