Skip to main content

Volume Monitor

The Volume Monitor at /data-quality/volume tracks row counts and table size over time, flags anomalies (empty / shrinking / explosively growing tables), and renders trend charts per table.

Snapshot vs. live​

Each catalog has a rolling history of volume snapshots. The page can:

  • Show current — query UC table stats live
  • Take snapshot — capture row counts and sizes into the _clxs_volume_snapshots table for trend analysis
  • Compare — current vs. last snapshot

Snapshots are how the trend charts and growth-rate badges are computed. Take a snapshot daily (via Scheduling) for meaningful trends.

Filter presets​

Quick-filter pills at the top:

  • All — every table
  • Empty — row_count == 0
  • Anomalous — flagged by the anomaly detector (Anomalies)
  • Growing — > 10% increase since last snapshot
  • Shrinking — > 10% decrease
  • Largest — top 10 by size
  • Smallest — bottom 10 by size

Datatable columns​

  • Table FQN
  • Current rows
  • Previous rows (last snapshot)
  • Change % (colour-coded)
  • Size in bytes
  • Trend icon (up / down / flat)

Charts​

Two chart types on the right rail:

  • Per-table line chart — row-count history when you select a row
  • Cumulative area chart — catalog-wide rows over time

Health score​

Each table gets a 0–100 health score combining:

  • Recency of last successful write
  • Conformance to its SLA row-count bounds
  • Anomaly severity

The score is also displayed on the DQ Scorecard.

API​

GET /data-quality/volume?catalog={catalog}&schema={schema}&table={table}

Returns one entry per table with current and previous counts, deltas, and trend metadata.