A quality-based framework for physical data warehouse design
Mokrane Bouzeghoub, Zoubida Kedad · 2000
Data warehousing is a software infrastructure which supports OLAP applications by providing a collection of tools which allow data extraction and cleaning, data integration and aggregation, and data organization into multidimensional structures which are suitable for decision making. At the design level, a data warehouse is defined as a hierarchy of view expressions. One key problem of the design of a data warehouse is the selection of a set of views to materialize in order to meet the performance requirements, that is, to minimize both the evaluation cost of the users queries and the maintenance cost of the materialized expressions. Existing approaches either propose algorithms that explore the search space by enumerating and evaluating all the possible solutions, or introduce heuristics that restrict the search space. In this paper, we propose a framework to support the physical design of a data warehouse; it is based on a logical model, called the view synthesis graph. A level of overlapping is defined to characterize the graph. This level allows to order the search space. The framework integrates some quality factors that can be used either to drive the exploration of the search space, or to reduce the search space, or both.