Running a query job interactively versus as a batch
BigQuery jobs run in one of two priority modes: interactive (default) or batch. Interactive queries execute as soon as possible and count immediately against concurrent rate limits, while batch queries are queued and scheduled to run when idle resources are available, making them ideal for non-urgent workloads. Choosing batch priority helps avoid quota contention and reduces slot competition with time-sensitive interactive queries.
1 · Learn the must-know
- Interactive queries (the default) are executed as soon as possible and are subject to concurrent rate limits per project.
- Batch queries are queued by BigQuery and run when idle resources are available, so they may wait but do not count toward the concurrent interactive query limit in the same way.
- You set query priority using the --batch flag in bq CLI, the priority field in the jobs.query/jobs.insert API, or the query priority setting in job configuration; it cannot be changed after submission.
- Batch queries have historically been capped at a maximum wait time (previously 24 hours) before automatically being promoted to interactive priority if not started.
- Batch priority does not reduce the cost of on-demand queries, pricing is based on bytes scanned regardless of interactive or batch mode.
- For flat-rate/capacity-based pricing (reservations), both interactive and batch jobs draw from the same assigned slot capacity, so priority mainly affects scheduling order, not slot allocation itself.
2 · Check your understanding
A retail company's data analysts run frequent interactive queries in BigQuery throughout the business day under on-demand pricing. A nightly job that recalculates a large sales aggregation table is not time sensitive, but when it overlaps with business hours it sometimes causes analysts to receive concurrent query quota errors. What should the data engineer change on the nightly aggregation job to resolve the contention without provisioning additional capacity?
What you have tried across GCP PDE's objectives, not a readiness score.
Designing data processing systems~22% of the exam0 of 17 tried
Ingesting and processing the data~25% of the exam0 of 11 tried
Storing the data~20% of the exam0 of 13 tried
Preparing and using data for analysis~15% of the exam0 of 11 tried
Maintaining and automating data workloads~18% of the exam0 of 15 tried
3 · Keep going
Ready for more? Take a weighted mock or try free practice questions.