Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»
The Nature of the Development Efforts


The first issue facing the data architect who must manage multiple data warehouse development efforts is the nature of the development efforts. Unless the data architect knows what kinds of efforts are occurring and how they relate to the overall architecture, he or she will have a very difficult time managing and coordinating them. Because the issues of development vary considerably depending on the approach, different types of data warehouse development efforts require very different approaches to management.


Multiple data warehouse development efforts occur in four typical cases, as outlined in Figure 6.17.


In the first, rare case shown in Figure 6.17, a corporation has totally separate and unintegrated lines of business for which data warehouses are being independently built by different development groups. The diverse lines of business report to the corporation, but other than sharing a company name, there is no business integration or sharing of data within the company. Such a corporate


d/w A    d/w B    d/w C    d/w D


line of business    line of business    line of business    line of business


ABCD


completely unintegrated lines of business each with their own data warehouses


northeast


midwest


western


southwest


d/w


d/w


d/w


d/w

the same data warehouse but with distributed parts



OLAP


different levels of data within the same data warehouse


different non-distributed parts of the detailed level of the data warehouse


Figure 6.17 Four possible meanings to “multiple groups building the data warehouse,» each interpretation very different from the others.

Скачать в pdf «Building the Data Warehouse»