Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


The output of this step is the identification of the major subjects that will make up the system, as well as their relationships to each other.


PARAMETERS OF SUCCESS: All major subjects are identified so that there are no conflicts in domain; they are identified at the highest level of abstraction.


One major parameter of success is that only primitive data be modeled. Another parameter of success is that the scope of the model be defined prior to starting the ERD modeling.


D2—DIS (Data Item Sets)


PRECEDING ACTIVITY: ERD definition.


FOLLOWING ACTIVITY: Performance analysis; data store definition.


TIME ESTIMATE: As long as one month per subject area.


NORMALLY EXECUTED ONCE OR MULTIPLE TIMES: Once for each subject area.


Each subject is further broken down-in terms of level of detail-into a dis (data item set). The dis contains attributes of data, the grouping of attributes, and keys. In addition, “type of” data is identified. Other structures of data here include connectors-representations of relationships-and secondary groupings of data. The output from this step is the fleshing out of the subject areas identified in D1.


PARAMETERS OF SUCCESS: All types of the major subject are identified; all connectors are correctly identified; all relationships are identified by a connector; all attributes are identified; all attributes are grouped with other attributes that share the same relationship to the key of the grouping of data; all multiply occurring groups of attributes are separated from singularly occurring groups of attributes; all recursive relationships are designed in the most general case necessary. Only primitive data is found here. Derived data is identified, stored, and managed elsewhere.

Скачать в pdf «Building the Data Warehouse»