A data engineering group already writes its nightly ETL as Python DAGs for an open-source scheduler. They want Google to run and patch that scheduler so they can keep the same DAG code. Which managed service should they use?
Show answer and explanation
Answer: B
Cloud Composer is a managed version of Apache Airflow, so teams can keep their existing DAG definitions while Google handles provisioning, upgrades and scheduler operations. Cloud Scheduler only triggers jobs on a cron timetable, Cloud Tasks queues asynchronous requests, and Workflows sequences service calls with its own definition syntax rather than running Airflow DAGs.
Why the other options are wrong
- A. Cloud Scheduler is a cron-style trigger that starts jobs at set times. It does not execute Airflow DAGs or manage task dependencies the way the team's existing code requires.
- C. Cloud Tasks is a queue for dispatching asynchronous requests between services. It has no DAG model, so it cannot host the team's existing nightly pipeline definitions.
- D. Workflows orchestrates calls between Google Cloud services using its own definition language. Porting the Python DAGs to it would require rewriting them, which the scenario explicitly wants to avoid.