Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


In Figure 6.20 a corporation has different sites in different parts of the world— one in the United States and Canada, one in South America, one in the Far East, and one in Africa. Each site has its own unique data, with no overlap of data— particularly of detailed transaction data—from one site to the next. The company wants to create a data warehouse for each of the disparate entities as a first effort in achieving an architected environment. There is some degree of business integration among the different organizations. At the same time, it is assumed that distinct business practices are carried on in each locale. Such an organization of corporate entities is common to many companies.


The first step many organizations make toward data warehousing is to create a local data warehouse at each geographical entity. Figure 6.21 shows the creation of a local data warehouse.


Each locale builds its own unique autonomous data warehouse according to its needs. Note that there is no redundant detailed data among the different locales, at least as far as transaction data is concerned. In other words, a unit of

Africa


d/w


Canada


d/w


Far East d/w


US


d/w


South America d/w


Figure 6.21 Local data warehouses are built at each of the autonomous operating divisions.


data reflecting transactions that belongs in Africa would never be found in the local data warehouse for Europe.


There are several pros and cons to this approach to building the distributed corporate data warehouse. One advantage is that it is quick to accomplish. Each local group has control over its design and resources. A sense of autonomy and control may make each local organization happy. As such, the benefits of the data warehouse can be proven throughout the corporation on a real-time basis. Within six months the local data warehouses can be up and running, and the organization at the local level can be deriving the benefits. The disadvantage is that if there is any commonality in the structure (not the content) of data across the organization, this approach does nothing to recognize or rationalize that commonality.

Скачать в pdf «Building the Data Warehouse»