Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


The different levels of data form a higher set of architectural entities. These entities constitute the corporate information factory, and they are described in more detail in my book The Corporate Information Factory, Third Edition (Wiley, 2002).


Some people believe the architected environment generates too much redundant data. Though it is not obvious at first glance, this is not the case at all. Instead, it is the spider web environment that generates the gross amounts of data redundancy.


Consider the simple example of data throughout the architecture, shown in Figure 1.11. At the operational level there is a record for a customer, J Jones. The operational-level record contains current-value data that can be updated at a moment’s notice and shows the customer’s current status. Of course, if the information for J Jones changes, the operational-level record will be changed to reflect the correct data.


The data warehouse environment contains several records for J Jones, which show the history of information about J Jones. For example, the data warehouse would be searched to discover where J Jones lived last year. There is no overlap between the records in the operational environment, where current information is found, and the data warehouse environment, where historical information is found. If there is a change of address for J Jones, then a new record will be created in the data warehouse, reflecting the from and to dates that J Jones lived at the previous address. Note that the records in the data warehouse do not overlap. Also note that there is some element of time associated with each record in the data warehouse.

Скачать в pdf «Building the Data Warehouse»