Features¶

dmp-af provides powerful features for running dbt at scale on Airflow.

Core Features¶

Domain-Driven Architecture¶

Organize models by domain into separate DAGs for parallel execution and data mesh architectures.

Flexible Scheduling¶

Run models on different schedules: hourly, daily, weekly, monthly, or on-demand.

Distributed Runs¶

Each dbt model becomes an independent Airflow task while preserving dependencies.

Manual Run DAG¶

Manually trigger arbitrary dbt runs for backfills and testing.

Advanced Features¶

Test Separation¶

Separate small, medium, and large tests into different execution paths.

Cross-Domain Dependencies¶

Models in different domains and schedules can depend on each other seamlessly.

Key Benefits¶

• Auto-Generated DAGs: No manual DAG writing required
• Idempotent Runs: Date intervals passed to every execution
• Team-Friendly: Analytics teams stay in dbt
• Enterprise-Ready: Multi-target, Kubernetes, and integration support