GCP Professional Data Engineer Practice Question

A BigQuery materialized view can store the pre-aggregated results (revenue, order count, and average order value by region, product, and channel). BigQuery refreshes the materialized view automatically and incrementally, typically within 30 minutes, so the dashboards continue to display recent data. Because the BigQuery optimizer can transparently rewrite incoming queries to use an eligible materialized view, no changes are required in the existing Looker Studio SQL. Query latency is reduced because the engine scans only the much smaller cached result instead of the 5-TB base table, and you pay only for the bytes processed in that cached result, lowering costs.

Partitioning and clustering the base table would cut scanned bytes for date-filtered queries, but every dashboard refresh would still scan all rows for the selected day and perform aggregations at runtime. Enabling query result caching helps only when the identical query text is rerun by the same user and expires after 24 hours; parameterized dashboard queries typically bypass this cache. Provisioning BI Engine memory can accelerate repeated queries, but it incurs an additional cost allocation and does not reduce the amount of data scanned from storage; it also does not guarantee the 30-minute freshness requirement without materializing aggregates.