QuestionQ98

VerifiedDesigning data processing systems

Your large e-commerce company relies on BigQuery for its data warehousing requirements. It has two main workload types:

  • Analyst dashboards: Interactive queries that business analysts run throughout the day and that require low latency for real-time insights.
  • ETL jobs: Large, complex batch queries that load and transform data overnight. These jobs are cost-sensitive, can tolerate longer execution times, and do not have the same performance requirements as the dashboards.

Analysts report inconsistent dashboard performance, especially during peak business hours. You must ensure dependable performance for both workload types while optimizing costs. What should you do?

Explanation

BigQuery reservations allocate slots into workload-specific pools and can isolate critical production workloads from competing ETL activity. An autoscaling reservation assigned to dashboard workloads supplies capacity as demand changes, while a separate ETL reservation or on-demand pricing lets the less time-sensitive, cost-sensitive batch workload use an appropriate cost model. BigQuery documentation identifies separate reservations and assignments as a way to isolate workloads and improve performance consistency.

Learn more

Community Discussion

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