Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


For all the autonomy of the lines of business, they are necessarily integrated at one level: the financial balance sheet. If the different lines of business report to a single financial entity, there must be integration at the balance-sheet level. In this situation, a corporate data warehouse might need to be built that reflects the corporate finances. Figure 6.19 shows a corporate financial data warehouse sitting above the different businesses.


The financial corporate data warehouse contains simple (and abstract) entities such as expenses, revenues, capital expenditures, depreciation, and the like. There is very little, if any, business data beyond that found on every balance sheet. (In other words, no attempt is made for a common corporate description


г

Africa    US    Canada    Far East South America


d/w    d/w    d/w    d/w    d/w


[____________________________________________1


Figure 6.20 Logically the same data warehouse.


of customer, product, sale, and so on in the financial data warehouse.) Of course, the data feeding the corporate financial data warehouse depicted in Figure 6.19 may come either from the “local” data warehouse or from the operational systems found at the individual operating company level.


Meta data is vital at the local level. With a corporate financial data warehouse, it is also needed at the corporate financial level. However, in this case, because no real business integration exists, there is no need to tie any of the meta data together.

Distributed Data Warehouse Development


Unlike the case of the unrelated warehouses, most businesses have some degree of integration among their disparate parts. Very few businesses are as autonomous as those depicted in Figure 6.19. A much more common scenario for the development of multiple data warehouse efforts is shown in Figure 6.20.

Скачать в pdf «Building the Data Warehouse»