SML Reference
The Semantic Modeling Language (SML) is an AtScale-provided modeling language that data engineers can use to create models within AtScale. This enables you to develop models programmatically and store them in an AtScale-connected Git repository. You can then deploy so that other users can connect to them via BI tools.
If you migrate to containerized AtScale from an installer-based version, you must migrate your existing models to SML. For more information, see Migration Considerations.
Data models created with SML are collections of objects, each of which defines a particular component of the model, such as a dataset or dimension. Objects are defined in YAML files and stored in your Git repository.
The following sections describe the different SML object types as well as the properties available for each:
- Catalog
- Package
- Calculations
- Connections
- Datasets
- Dimensions
- Metrics
- Models
- Composite Model
- Row Security
Migrated models may contain additional properties not documented here. These properties are required to ensure all aspects of the original model functions properly in SML. You should not include these properties in new models created with SML.