Data Warehouse Query Cost Allocation

Distribute cloud data warehouse spend at the query level by combining runtime-derived compute credits with optional cloud services fees and cache relief.

Compute credits consumed by the virtual warehouse each hour of runtime.
Elapsed execution time from query start to completion in seconds.
Blended contract price per compute credit billed by your provider.
Defaults to 1.0. Use >1 when workload scaling or isolation boosts credit draw for this query.
Defaults to 0 GB. Enter logical data volume scanned to include cloud services charges.
Defaults to $2.25 per TB scanned when left blank.
Defaults to 0%. Values above 1 are treated as percentages.

For FinOps scenario planning; reconcile with provider invoices before chargeback.

Examples

  • 4 credits/hour, 180 s runtime, $3.25/credit, concurrency 1.05, 350 GB scanned, 20% cache hit ⇒ $1.30 per query
  • 16 credits/hour, 75 s runtime, $2.70/credit, default concurrency, 1,200 GB scanned ⇒ $4.18 per query

FAQ

How should I estimate the concurrency factor?

Use observed warehouse scaling multipliers or query-level resource monitors. When an isolation policy or queue bump doubles the virtual warehouse size, enter 2.0 so compute credits reflect that amplification.

Do I need to include storage or stage charges?

This model isolates per-query compute and cloud services fees. If you amortise storage across workloads, add it outside the query calculation to avoid double counting.

What units should data scanned use?

Provide gigabytes as reported by the warehouse. The calculator converts to terabytes before applying the optional services rate.

Can cache hits reduce compute cost?

Result caching suppresses cloud services charges, but compute credits were still consumed during execution. Adjust the concurrency factor if cache hits allowed you to downsize the warehouse.

Additional Information

  • Runtime is converted to hours before multiplying by credits per hour.
  • Services cost prorates optional GB input to TB using the provided or default rate.
  • Result cache hit rate trims the services charge by the surviving uncached share.
  • Output is rounded to two decimals and expressed in USD per query.