Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»

The data warehouse is capable of holding much more than internal, structured data. There is much information relevant to the running of the company that comes from sources outside the company.

External data is captured, and information about the meta data is stored in the data warehouse meta data. External data often undergoes significant editing and transformation as the data is moved from the external environment to the data warehouse environment. The meta data that describes the external data and the unstructured data serves as an executive index to information. Much can be done with the index information, such as placing it in both disk storage and near-line storage, creating a link between the data warehouse and unstructured data, doing internal index processing, and so forth. In addition, a “notification” service is often provided whenever a new entry is made into the data warehouse.

External and unstructured data may or may not actually be stored in the data warehouse. By associating external and unstructured data with a data warehouse, the organization precludes the need to store the external and unstructured data in multiple places. Because of the bulk of data that is associated with unstructured data, it is best to store at least part of the unstructured data on a bulk storage medium such as near-line storage.

Any architecture that must be implemented all at once, in a big bang, is doomed to failure in today’s world. There simply is too much risk and too long a period to wait until there is a payback. In addition, trying to freeze changes to consider any path that is revolutionary, rather than evolutionary is unrealistic.

Скачать в pdf «Building the Data Warehouse»