Each business intelligence project requires a data model, which defines the fields available in the project and their relationships. GoodData simplifies the data modeling process by separating the model into two components: the physical data model and the logical data model.
The physical data model (PDM) is the set of tables used to store data in the project; in GoodData, the PDM is entirely created and managed by the platform.
A GoodData user or developer never needs to interact with the physical data model, which is created and managed entirely within the GoodData platform. Therefore, they can concentrate on the data relationships in the model and leave the details of database structure to the GoodData Platform.
Instead, you build the logical data model (LDM), which requires only that you specify the datasets of your project, including their fields, and then build the connections between them.
The logical data model is built in the CloudConnect LDM Modeler, a visual interface that is an integrated component of CloudConnect Designer. When finished, your LDM can be published to your projects directly from the LDM Modeler interface.
- Effective data modeling requires a distinct set of skills that may not be part of a general software engineering background. If you are unsure if you or your team has the appropriate skills, please contact GoodData Customer Support.
- When you make changes to your logical data model in CloudConnect Designer, the updated model must be republished to any project in GoodData that utilizes the model. Otherwise, the changes that you made locally are not reflected in the project.
- Publication of a logical data model update, which may include the removal of facts or attributes, causes the automatic removal of any metric, report, or dashboard that references any of the removed objects. Other project objects that reference a removed metric, report, or dashboard are also removed, resulting in a cascading deletion of objects throughout the project. High-impact changes to your LDM should be applied with caution and always in a testing project first. For more information, see Data Modeling and Logical Data Model.