Cost of keeping incoming data query-ready

FOR CONTEXTThe work included in fresh-data-path costFresh-data pathQuery-ready dataNewlywrittenOrderedraw dataOrdered pre-aggregatedFreshdatacontinuouslyQuery engineQuerycomputeFastanswerscontinuously1Store incolumns2Orderdata3Order and pre-aggregate data
Drag to explore

Accumulated fresh-data-path cost

Accepted active-ingestion pair: redshift-t2-20260812-SUPER. Fresh-data-path cost uses the accepted modeled 32 active writer RPUs for complete producer uptime plus MSK, charged once. Reader query cost uses committed hourly RPU-second allocations at $0.467/RPU-hour in eu-west-2; billed_times is the historical field name, not invoice evidence. Storage, client cross-AZ transfer, producer VM, discounts and post-ingestion observations excluded. Accepted SUPER read-path comparison through the 113,219,565,734-row reference endpoint; existing 190 aggregate and 33 SUPER drill-down observations and their matched ClickHouse files retained. Query curves are step sums. Complete-ingest preparation cost is allocated proportionally by row progress; it is not a metered time series. Zero is retained in readouts and omitted on logarithmic axes.