A conceptual solution for representing time in data warehouse dimensions

摘要

Data Warehouses (DWs) use an omnipresent time dimension for keeping track of changes in measure values. However, this dimension cannot be used to model changes in other dimensions. On the other hand, Temporal Databases (TDBs) have been successfully used for modelling time-varying information. Bringing together these two research areas, leading to Temporal Data Warehouses (TDWs), provides the necessary solutions for managing time-varying data in dimensions. In this paper, we introduce temporal extensions for the MultiDimER model, a conceptual multidimensional model. In our model we allow the inclusion of valid and transaction time, which are obtained from source systems, in addition to the data warehouse loading time. Our model allows a conceptual representation of time-varying levels, attributes, and hierarchies. For the latter, we discuss different cases depending on whether the changes in levels affect the relationships between them.