Integration and Reuse of Heterogeneous Information: Hetero-Homogeneous Data Warehouse Modeling in the Common Warehouse Metamodel

The corporate data warehouse integrates data from various operational data stores of a company. These operational data stores may be heterogeneous with respect to the represented information. The hetero-homogeneous data warehouse modeling approach overcomes issues associated with the integration of heterogeneous information from the operational data stores by featuring a generally homogeneous schema which may be interspersed with heterogeneities in well-defined portions of the data. In order to leverage the capabilities of existing business intelligence (BI) tools for the analysis of hetero-homogeneous information, the schema must comply with the metamodel of the particular BI tool. The Common Warehouse Metamodel (CWM) is a standard for data warehouse metadata which facilitates the reuse of data across multiple BI tools. In this paper, we present guidelines for modeling hetero-homogeneous data warehouses in the CWM. We demonstrate feasibility with a proof-of-concept prototype for the model-driven implementation of hetero-homogeneous data warehouses.