Creating a Data Model

How you work with a data model depends on the type of space you use. For the most powerful features, use an Advanced space and create the data model yourself.

For Automatic spaces, Birst automatically creates the data model and hierarchies. For Discovery spaces, Birst creates the hierarchies, levels, and keys for you, and you need to create joins and process the data. See About Discovery Spaces.

Data Models and Advanced Spaces

For Advanced spaces, you can specify the data model settings yourself. You select which sources to enable, set the primary keys, define the grain of your data sources, and set the properties of the columns in your sources.

You can also create hierarchies and relate these hierarchies to the incoming data sources. When Birst understands how hierarchies relate to incoming data, it can build a very sophisticated analytical data structure.

The data model settings determine the dimensional model that Birst generates from your data sources when you process the data.

An alternative to creating your own data model for an Advanced space is to use the Update Warehouse Model data model automation tool. Using this tool, Birst derives a dimensional data model for your source data based on the key columns that make the data set unique. Using this tool, you do not have to create your own hierarchies and grains. All modeling is performed from the Data Flow page. See About Update Warehouse Model.

Birst includes a Model Scan function to validate your data model after creating it or making changes to it.

Prior to Creating a Data Model

Upload your data sources using Upload Files, Birst Connect, Live Access, or an Application Connector.

High-Level Steps for Creating a Data Model for a Classic Advanced Space

  1. Enable the sources.
  2. Set or modify the primary key for each source.
  3. Specify the joins between sources.
  4. Define hierarchies and levels.
  5. Define the grain of your data sources.
  6. Define column properties.
  7. Run a model scan to validate your data model.
  8. Process your data.

High-Level Steps for Creating a Data Model for a Classic Advanced Space Using Update Warehouse Model

  1. Enable the sources.
  2. Set or modify the primary key of your sources.
  3. Create joins between your sources.
  4. Click the Update Warehouse Model button.
  5. Run a model scan to validate your data model.
  6. Process your data.

See Also
Creating Hierarchies
Defining the Grain of a Data Source
Defining Column Properties
About Discovery Spaces
About Update Warehouse Model
Validating Your Data Model
Comparing and Merging Spaces