Amazon Managed Workflows for Apache Airflow
Managed Apache Airflow
MWAA runs Apache Airflow for you to orchestrate data pipelines with DAGs written in Python; AWS handles the scheduler, workers, web server, and upgrades.
- Cost$$ Medium
- Difficulty●●●●●
- Costs money when idle?Yes
- Managed by AWS
- In examsSAA
Think of it asLike hiring a professional conductor to lead your data orchestra from the score you wrote.
When to use it
✓ Use it when
Teams already familiar with Airflow or that need to orchestrate many complex, interdependent data jobs.
✗ Avoid it when
Simple workflows mostly using AWS services → Step Functions.
Common uses
- Orchestrate daily Glue + EMR + Redshift ETL
- Move self-hosted Airflow to AWS
- Schedule periodic model training
What does it cost?
- Pricing model: Charged per hour of environment runtime (by size), additional workers/schedulers, and metadata database storage.
⚠️ Cost traps
The environment is billed continuously even when no DAGs are running.
