Understand Logical Data Model


What You Learned

In this course, you learned that:

  • The LDM is a part of the semantic layer in GoodData and defines entities and their relationships.
  • It is a way to abstract the physical data structure and allow for easier report creation.
  • It also enables ad-hoc reporting both for data analysts and users less skilled in data analysis.
  • Each GoodData workspace has an LDM.
  • Data on a server need to be stored in a format defined by the LDM of each workspace.
  • The LDM in GoodData is represented as a graph of boxes (datasets) connected by arrows (relationships).
  • The LDM consists of several types of objects:
    • Datasets
    • Relationships between datasets
    • Attributes (and their labels)
    • Facts
    • Date dimensions
    • References
    • Primary keys (single attribute or compound of attributes and references)
  • Part of the GoodData application is the modeler tool, which allows you to view and modify the LDM of a particular workspace. To get to the modeler tool, click the “Data” tab.
  • The objects from the LDM are used in other parts of GoodData such as Analytical Designer or Metric Editor.
  • The LDM defines what you can and cannot analyze in GoodData.
  • You can slice and filter by anything from the same dataset as the fact you’re using.
  • Measures based on facts or attributes from one dataset can be sliced and filtered by attributes from another dataset if you can get from the second dataset to the first one following the direction of the arrows.

What’s Next

To learn how to design logical data models for GoodData, review the Designing Data Models course.

You can also always refer to our documentation and discuss the logical data model in our community.