A reminder that you don't necessarily need Airflow / Composer to orchestrate data in and out of BigQuery. If your workflows are not that complex or numerous, take a look at Google Cloud Workflows.

It's a serverless orchestration engine, with good integration with other GCP services, but you can also call external APIs.

Some advantages I see:
- serverless, no servers to manage
- your workflows stored as code, manage Infra as Code with Terraform
- interact with external APIs
- good connectors to interact with other GCP services
- straightforward pricing: at the time of writing: 5k steps free per month, 0.01 USD per 1k steps after, external API call are priced separately (2k/mo free, 0.025 USD per 1k after)

I've used it before to load data from Google Cloud Storage into BigQuery tables on a schedule and perform other simple tasks. It gets the job done with no fuss. You can of course schedule the Workflow to run with Cloud Scheduler.

See my previous post for a short walk through showcasing how BQ and Workflows play together.

Found it useful? Subscribe to my Analytics newsletter at notjustsql.com.