Introduction to Data Modeling Deliverables
The deliverables of the data modeling process serve as essential documentation and reference points. These components ensure clarity, accuracy, and traceability throughout the lifecycle of a data model.
Key Deliverables in the Data Modeling Process
Diagrams
A data model contains one or more diagrams. The diagram is a visual that captures the requirements in a precise form. It depicts a level of detail (e.g., conceptual, logical, or physical), a scheme (relational, dimensional, object-oriented, fact-based, time-based, or NoSQL), and a notation within that scheme (e.g., information engineering, unified modeling language, object-role modeling).
Definitions
Definitions for entities, attributes, and relationships are essential to maintaining the precision of a data model.
Issues and Outstanding Questions
Frequently, the data modeling process raises issues and questions that may not be addressed during the data modeling phase. In addition, often the people or groups responsible for resolving these issues or answering these questions reside outside of the group building the data model. Therefore, often a document is delivered that contains the current set of issues and outstanding questions. An example of an outstanding issue for the student model might be, “If a Student leaves and then returns, are they assigned a different Student Number or do they keep their original Student Number?”
Lineage
For physical and sometimes logical data models, it is important to know the data lineage, that is, where the data comes from. Often, lineage takes the form of a source/target mapping, where one can capture the source system attributes and how they populate the target system attributes. Lineage can also trace the data modeling components from conceptual to logical to physical within the same modeling effort. There are two reasons why lineage is important to capture during Data Modeling. First, the data modeler will obtain a very strong understanding of the data requirements and, therefore, is in the best position to determine the source attributes. Second, determining the source attributes can be an effective tool to validate the accuracy of the model and the mapping (i.e., a reality check).
