You need to take streaming data from thousands of Internet of Things (IoT) devices, ingest it, run it through a processing pipeline, and store it for analysis. You want to run SQL queries against…

Question

Options

• ACloud Dataflow, Cloud Pub/Sub, BigQuery
• BCloud Pub/Sub, Cloud Dataflow, Cloud Dataproc
• CCloud Pub/Sub, Cloud Dataflow, BigQuery
• DApp Engine, Cloud Dataflow, BigQuery

• A
  8% (5)
• B
  5% (3)
• C
  71% (44)
• D
  16% (10)

Explanation

Cloud Pub/Sub → Cloud Dataflow → BigQuery is the canonical GCP streaming pipeline: Pub/Sub acts as a scalable message broker that decouples thousands of IoT devices from your processing layer, Dataflow runs the stream processing transformations, and BigQuery stores the results in a fully managed data warehouse that accepts SQL queries.

Why the distractors fail:

• A reverses the first two services - you can't run Dataflow processing before the data has been ingested into Pub/Sub.
• B uses Cloud Dataproc (a Hadoop/Spark cluster service built for batch jobs) instead of BigQuery, which doesn't natively support SQL analytics the way BigQuery does.
• D replaces Pub/Sub with App Engine, which is an application-hosting platform, not a message ingestion service for IoT device streams.

Memory tip: Think "Pub → Flow → Big" - Publish the raw stream, Flow it through transformations, then land it in BigQuery for big analysis.

No community discussion yet for this question.