Airflow pricing at a glance
Apache Airflow is Apache-2.0 licensed software for programmatically authoring, scheduling and monitoring workflows. Workflows are defined in Python; the scheduler runs tasks on workers while following their dependencies. Its web application supports monitoring workflow and task status, logs, and troubleshooting.
What the software includes
Airflow provides operators for Google Cloud, AWS, Azure and other third-party services, and users can define or extend operators for their environment. It is distributed through PyPI, Docker images and source releases; these are installation methods, not subscription tiers.
Build a deployment budget
Model the operating setup you intend to run: the scheduler and workers, plus a production database. Airflow documents PostgreSQL and MySQL support and says SQLite should not be used in production. For high-volume, data-intensive tasks, the project recommends delegating work to specialized external services rather than passing large quantities of data between tasks. Include those execution services with the orchestration infrastructure in the budget.
Account for maintenance
Use the project's version-specific constraint files for repeatable PyPI installations: it warns that an unconstrained install can sometimes fail or be unusable. Provider packages use their own semantic versioning and can set minimum Airflow-version requirements. Plan installation checks, dependency testing and upgrades as operating work.
Our take
Choose Airflow when you want to operate the scheduler, workers and production database for Python-defined workflows, and can own the related installation and upgrade work. Compare options against the same workflow and execution responsibilities; the license does not determine the operating bill.