Pick a topic. Pick an activity. Go.

The best way to learn something is to try it yourself. So we've turned a bunch of the demos our instructors do into Do-It-Now activities, where you can try each concept out yourself in quick, bite-size pieces.

Each topic below opens a set of independent activities. They are not a tutorial and are not meant to be worked in order — every activity stands on its own and illustrates a single concept, so start anywhere and do as many or as few as you like.

If you are using these as part of an Instructor Led Training (ILT), make sure to listen to your instructor and coordinate your efforts with their direction.

Topics

BigQuery BigQuery Do-It-Now Activities Query at scale, then take the machinery apart: caching, joins, nested and repeated data, partitioning and clustering, execution details, and loading. 30 activities Dataflow Dataflow Do-It-Now Activities Apache Beam pipelines on a managed runner — batch and streaming, the job graph, and what autoscaling actually does under load. Activities coming soon Data Engineering Misc. Data Engineering Do-It-Now Activities The rest of the platform: orchestration with Composer, governance and profiling with Dataplex, Spark on Dataproc, open table formats, and data protection. Activities coming soon

Before you start

Cost

Many of these activities cost money to run. Every activity shows its cost under the title:

🟢 No cost Free to run.
💰 Low cost Small charges.
💸 Higher cost Scans large tables; generates significant charges.

Sample data

Activities note the data they need in their own setup section. Most BigQuery activities use the roi-bq-demos.bq_demo dataset; some rely on additional tables derived from it, which you can generate using the directions here.

There is also a much smaller dataset that works with the demos: roi-bq-demos.bq_demo_small. Queries are much faster against it, so it isn't as effective at illustrating query speed benefits.