Skip to main content
Paradime supports dbt™ Mesh, so teams can split work across several dbt™ projects and workspaces and still reference each other’s models. In workspaces on dbt™ 1.7 to 1.12, Paradime resolves cross-project references with dbt-loom, which reads each producer project’s public models from one of its Bolt schedules.

​
Key components

  • Cross-project references: the foundation of a multi-project setup. A consumer project references a producer project’s public models with a two-argument {{ ref() }}.
  • Model groups: organize nodes in your dbt™ DAG that share a logical connection (for example, a functional area) and assign an owner to the whole group.
  • Model access: control which models other groups and projects can reference.

​
Paradime dbt™ Mesh features

​
Lineage across workspaces

Catalog lineage crosses one workspace boundary in each direction. In a consumer workspace, the producer models you reference appear in lineage, tagged with the workspace they come from. In a producer workspace, a model’s lineage shows the models in other workspaces that reference it. See dbt™ Mesh projects.

​
Paradime Data Catalog

The producer models that a consumer project references appear in the consumer workspace’s Catalog, tagged with the workspace they come from.

​
Paradime Bolt

Use the On run completion trigger to start a consumer schedule when a producer schedule finishes, even across workspaces. dbt-loom also reads the producer’s public models from one of its Bolt schedules, so keep that schedule running.

​
Column-Level Lineage Diff

Column-Level Lineage Diff shows how a pull request in a producer project affects columns in its consumer projects, so you can spot breaking changes before you merge and tell the teams that own the affected models.