Best practices for cross-model references in diagrams

When you copy data objects from a diagram into a diagram in another data model, a cross-model reference is created. This document outlines some best practices for creating cross-model references without affecting the performance of the workbench.

Many data modelers choose to create multiple smaller data models to help componentize their work. Data models are often used to model separate processes, but sometimes, these processes overlap. In order to better understand these processes and how they work together, you can create diagrams that use data objects from more than one data model.

When these types of diagrams are created, cross-model references are also created. In order to view or modify these data objects, the referenced data model is also opened when you open another model that is cross-referenced. If the source model is deleted while the model with references to it remains closed, then the references to the source model are considered null objects in the Data Project Explorer, and the view of the objects on the diagram also becomes invalid.

Cross-model references are created in the following scenarios:

In order to improve workbench performance, you can create these data models that model different business processes. Then, when you are ready to visualize how each of these models works together, create a separate model that contains diagrams only. By using this method, you can continue to work on each of the data models separately, updating as necessary. You can improve workbench performance even more by closing all other models and working on a single data model at a time. All changes will appear in the diagramming model when you are ready to use the diagramming model again.


Feedback