QuestionQ71

VerifiedDesigning data processing systems

You have two projects that run BigQuery jobs:

  • One project runs production jobs with strict completion-time SLAs. These high-priority jobs must have the required compute resources available when needed. They generally never fall below 300-slot utilization, but occasionally spike by an additional 500 slots.
  • The other project is for users to run ad-hoc analytical queries. This project generally never uses more than 200 slots at a time. You want these ad-hoc queries billed according to the amount of data users scan, rather than slot capacity.

You need to make sure that both projects have the appropriate compute resources available. What should you do?

Explanation

A dedicated Enterprise Edition reservation isolates the SLA workload and supplies its always-needed 300 baseline slots, while autoscaling can add up to 500 slots for occasional demand spikes. Keeping the ad-hoc project on on-demand billing charges its queries by data processed rather than by reserved slot capacity. BigQuery reservations are intended for capacity-based billing and workload isolation; on-demand compute is billed per TiB scanned.

Learn more

Community Discussion

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