QuestionQ49

VerifiedPlanning and implementing a cloud solution

You are building a data lake on Google Cloud for your Internet of Things (IoT) application. The IoT application has millions of sensors that are constantly streaming structured and unstructured data to your backend in the cloud. You want to build a highly available and resilient architecture based on Google-recommended practices. What should you do?

Explanation

Streaming IoT sensor data to Pub/Sub provides reliable message queuing and decoupling of ingestion from processing. Dataflow consumes from Pub/Sub and streams data to Cloud Storage, which accommodates both structured and unstructured data in a data lake architecture. This is Google's recommended pattern for high-volume IoT ingestion. Storage Transfer Service (B, D) transfers files between object and file storage systems such as Cloud Storage and cannot write a stream into BigQuery, and devices cannot stream directly into Dataflow without an ingestion layer such as Pub/Sub (C, D). Dataprep (C) is a data-preparation tool, not a streaming ingestion path.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!