Dimension - Type 1

Submitted by wolf on

Dimension Type 1


Flattened structure that stores attribute information for one class of data (e.g. person, place, entity). In simple terms, dimensions give who, what, where of a fact. Limited A2B Data control attributes are added to record last update information.

General examples of Type 1 dimensions:  Shift, Department, US States, Zipcodes

Type 1 dimensions tend to hold information that does not change often like a list of territories or job codes. The user attributes are overwritten with the current values when changes occur.  In the below example, the Territory Name, Region Code and Group are the user attributes with the ID being the natural key.

    Territory dimension type 1       

Data Change Observation

Data  changes are observed for a given Natural Key(NK).  New NK combinations will generate a new record.  Existing NK combinations will update the record with most recent values. If the NK combination no longer exists in the source system, the row is marked as deleted. No history of data changes is retained.

The New Record will have attributes set to the following values:
          Delete Row = 'N'
          Insert DateTime = Processed Date/Time
          Update DateTime = Processed Date/Time

The Updated Record will have the following attribute values set:
          Update DateTime = Processed Date/Time

The Deleted Record will have the following attribute values set:
          Delete Row = 'Y'
          Update DateTime = Processed Date/Time

Primary Key

Primary Key is assigned for the unique combination of Natural Key of the Driver Table.