Pre-aggregation lag

FOR CONTEXTPre-aggregation lag depends on how the data is kept freshFresh-data pathQuery-ready dataNewlywrittenOrderedraw dataOrdered pre-aggregatedFreshdatacontinuouslyQuery engineQuerycomputeFastanswerscontinuouslyPOSSIBLE LAG
Drag to explore0.0B / 100B

Lag between raw data and pre-aggregated data

Databricks: time between observed completed refreshes. ClickHouse: insert-time incremental updates. Intervals between observed completions. The compact freshness history may omit intermediate completions; no missing completion timestamps are invented.