Skip to main content

Storage Costs

The Storage Costs page at /finops/storage breaks down lakehouse storage spend by catalog, schema, table, and storage tier. It tells you which tables are eating the bill and where to look for optimisation.

What's measured​

  • Active storage — current Delta table sizes
  • Time-travel storage — historical Delta versions retained per delta.deletedFileRetentionDuration
  • Cloud blob class — Hot / Cool / Archive (for Azure), Standard / IA / Glacier (for AWS), etc.
  • Compression ratio — uncompressed-source vs. on-disk size

Cards​

  • Total storage GB
  • Total monthly cost
  • Time-travel as % of total (high = candidate for VACUUM)
  • Top catalog by cost

Breakdown table​

Per row:

  • Catalog / schema / table FQN
  • Active GB
  • Time-travel GB
  • Total GB
  • Monthly cost (effective rate × GB)
  • Storage tier
  • Last modified

Sort by cost descending. Select multiple rows for bulk actions:

  • Vacuum now — runs VACUUM to clear time-travel beyond retention
  • Optimize now — runs OPTIMIZE to compact small files
  • Move to colder tier — for tables not accessed recently

Filters​

  • Catalog / schema scope
  • Min size threshold (e.g. only show tables > 1 GB)
  • Last accessed before — surface stale data
  • Storage tier

Recommendations​

The page surfaces auto-generated recommendations (also viewable on the Recommendations page):

  • Tables with > 50% time-travel where retention is unused
  • Tables with high small-file count (OPTIMIZE candidates)
  • Tables not read in 90+ days (archive candidates)

Each recommendation shows estimated savings.

API​

GET /finops/storage?catalog=...&min_gb=1
POST /finops/storage/vacuum   { "table_fqns": [...] }
POST /finops/storage/optimize { "table_fqns": [...] }