Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»

When an organization takes a short-term approach, justifying the data warehouse is hard to do. The long-term costs of building many multidimensional database environments is very high. When an organization takes a long-term view and builds a data warehouse, the long-term total cost of data warehousing and data marts drops significantly.

Data Warehousing across Multiple Storage Media

One interesting aspect of a data warehouse is the dual environments often created when a large amount of data is spread across more than one storage media. One processing environment is the DASD environment where online/ interactive processing is done. The other processing environment is often a tape or mass store environment, which has essentially different features. Logically, the two environments combine to form a single data warehouse. Physically, however, the two environments are very different. In many cases, the underlying technology that supports the DASD environment is not the same technology that supports the mass store environment. Mixing technologies in the data warehouse environment is normal and natural when done this way.

However, there is another way that technology can be split that is not normal or natural. It is conceivable that the data warehouse environment—the DASD portion—is split over more than one technology. In other words, part of the DASD-based data warehouse resides on one vendor’s technology and another part of the data warehouse resides on another vendor’s database technology. If the split is deliberate and part of a larger distributed data warehouse, such a split is just fine. But if the split occurs for political or historical reasons, splitting part of a data warehouse onto different vendor platforms is not advisable.

Скачать в pdf «Building the Data Warehouse»