Associate Data Practitioner free question 8 / 10 · Domain 3: Data Pipeline Orchestration
Once a day, you want to run one SQL statement in BigQuery to update a summary table. There are no complex external dependencies. Which is the minimal-setup candidate?
- Your own cron job on Compute Engine
- A scheduled query configured in BigQuery
- A dedicated Cloud Composer environment
- A long-running Dataproc Spark cluster
Show the answer and explanation
Answer: B
Why this is correct
The BigQuery page on scheduling queries says: "You can schedule queries to run on a recurring basis." It manages running SQL on a schedule, so it meets the requirement without standing up a dedicated orchestration environment.
Why the other options are wrong
"Your own cron job on Compute Engine" … It can be done, but you would have to manage the availability and maintenance of the VM, which is more to build than a managed scheduled query.
"A dedicated Cloud Composer environment" … It can be done, but for a single SQL statement it adds the operation of an Airflow environment, so it is not the minimal setup.
"A long-running Dataproc Spark cluster" … A distributed Spark platform is not needed for the goal of running SQL on a schedule in BigQuery, and it makes the setup larger.
Key point
For small requirements, start by considering a dedicated managed feature. When dependencies between steps or integrations with other services grow, reassess whether you need broader orchestration.