Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


NORMALLY EXECUTED ONCE OR MULTIPLE TIMES: Once.


DELIVERABLE: Granularity analysis.


Once the model has been analyzed and brought up to a level of sufficient quality, the next step is to do breadbox analysis. Breadbox analysis is a sizing-in terms of gross estimates-of the DSS environment. If volume of data is going to be a problem, it is important to know that at the outset. Breadbox analysis simply projects-in raw terms-how much data the data warehouse will hold.


The output of breadbox analysis is simple-if the data warehouse is to contain large amounts of data, then multiple levels of granularity need to be considered. If the data warehouse is not to contain a massive amount of data, then there is no need to plan the design for multiple levels of granularity.


PARAMETERS OF SUCCESS: An estimate of the amount of data-in terms of number of rows-both on the one-year horizon and on the five-year horizon for the entire data warehouse environment, is the result of the process. Based on the results of the estimate, the issue of whether different levels of granularity are needed is decided. If multiple levels of granularity are needed, defining exactly what those levels are is a part of the output of this step.


DSS3—Technical Assessment


PRECEDING ACTIVITY: The commitment to build a data warehouse. FOLLOWING ACTIVITY: Technical environment preparation.


TIME ESTIMATE: One week.


NORMALLY EXECUTED ONCE OR MULTIPLE TIMES: Once.


DELIVERABLE: Technical environment assessment.


The technical requirements for managing the data warehouse are very different from the technical requirements and consideration for managing data and processing in the operational environment. That is why a separate, central store of DSS data is so popular.

Скачать в pdf «Building the Data Warehouse»