Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


There is, then, a very real case for storing summary data as well as detailed data. DSS and EIS processing ought to make as much use of summary data as they do of detailed data. Summary data is much less voluminous and much easier to manage than detailed data. From an access and presentation perspective, summary data is ideal for management. Summary data represents a foundation on which future analysis can build and for which existing analysis does not have to be repeated. For these reasons alone, summary data is an integral part of the DSS/EIS environment.


Keeping Only Summary Data in the EIS


Some very real problems become evident with keeping just summary data. First, summary data implies a process—the summary is always created as a result of the process of calculation. The calculation may be very simple or complex. In any case, there is no such thing as summary data that stands alone— summary data of necessity stands with its process. To effectively use summary data, the DSS analyst must have access to and an understanding of the process that has been used to shape it. As long as DSS and EIS understand this relationship between process and summary data and as long as EIS and DSS can profitably use the summary data that has resulted from the process of calculation, then summary data constitutes an ideal foundation for EIS and DSS. However, if the analysts that are doing EIS/DSS analysis do not understand that


process is intimately related to summary data, the results of the analysis can be very misleading.

Скачать в pdf «Building the Data Warehouse»