Practice and extend - Step 13 of 15
Run locally, then choose automation
Practice local pipeline runs, then compare common ways to schedule and deploy them.
Finish the course with local runs
This course focuses on building confidence with a pipeline on your own machine. Before someone else or a scheduler runs it, make sure you can validate it, run a small date range, and understand the output.
Run the same local checks you used to prove the pipeline:
bruin validate commerce/pipeline.yml
bruin run commerce/pipeline.yml \
--start-date 2026-01-01 \
--end-date 2026-01-03
Use a date range you understand and inspect the result after each run. Choosing the start and end dates yourself makes it easier to check that the pipeline changed the intended data.
Ways to automate runs and deploy
Once a pipeline works locally, a team can place the project somewhere that runs it on a schedule. These are common options. You do not need to set up any of them for this course.
Local cron
- What: A timer on your own computer runs a command at a chosen time.
- How: Add a cron entry that calls
bruin runwith the date range you need. - Why: It is a simple way to experiment with scheduled local runs.
GitHub Actions
- What: A scheduled job runs from the repository on GitHub.
- How: A workflow checks out the project and runs the Bruin CLI with stored connection settings.
- Why: The schedule and run instructions live alongside the pipeline code.
Cron on a server
- What: A timer on a server runs the pipeline instead of your laptop.
- How: Install the Bruin CLI on the server and have cron run the same command on a schedule.
- Why: The run can continue when your computer is off.
Bruin Cloud
- What: A managed service runs and observes Bruin projects.
- How: Connect the project, configure its connection settings, and choose a schedule in Bruin Cloud.
- Why: It provides shared run history and scheduling without maintaining a server yourself.
The option changes where and when the pipeline runs. The local validation, date range, SQL, and data checks you practiced in this course stay the same.
Checkpoint
Run validation and a small date range locally. You should be able to explain which dates ran, what changed, and which automation option you would explore next. Do not create a workflow, configure a schedule, or deploy the project for this course.