BigQuery Data and Compute Nodes

Question

Does Google BigQuery use different nodes for storing data and for computation of queries? I have seen that Amazon Redshift uses nodes that do both and Snowflake has a patent-pending architecture that separates storage and compute layers.

Thank you for the help.

Answer 1

Yes, the storage and compute layers are separate with Google BigQuery. You don't need to manage either layer with BigQuery since it is considered a serverless architecture. Storage automatically scales with the data you put into it, and compute automatically scales with the needs of the query.

If you are interested in learning more about how BigQuery works, you can check out the Google Research paper on Dremel.