Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»

Granularity of data can be raised in many ways, such as the following:

■■ Summarize data from the source as it goes into the target.

■■ Average or otherwise calculate data as it goes into the target.

■■ Push highest/lowest set values into the target.

■    Push only data that is obviously needed into the target.

■    Use conditional logic to select only a subset of records to go into the target.

The ways that data may be summarized or aggregated are limitless.

When building a data warehouse, keep one important point in mind. In classical requirements systems development, it is unwise to proceed until the vast majority of the requirements are identified. But in building the data warehouse, it is unwise not to proceed if at least half of the requirements for the data warehouse are identified. In other words, if in building the data warehouse the developer waits until many requirements are identified, the warehouse will never be built. It is vital that the feedback loop with the DSS analyst be initiated as soon as possible.

As a rule, when transactions are created in business they are created from lots of different types of data. An order contains part information, shipping information, pricing, product specification information, and the like. A banking transaction contains customer information, transaction amounts, account information, banking domicile information, and so forth. When normal business transactions are being prepared for placement in the data warehouse, their level of granularity is too high, and they must be broken down into a lower level. The normal circumstance then is for data to be broken down. There are at least two other circumstances in which data is collected at too low a level of granularity for the data warehouse, however:

Скачать в pdf «Building the Data Warehouse»