slashpeak
Solutions

Query waste

Most bills come from a few queries that look filtered in SQL and still read everything. We rank them by monthly cost and show the filter, cluster key or column list that would have skipped the rest.

Get my free BigQuery scanView all products

On BigQuery

Available

Partition filters from job_stages, unpartitioned date scans, SELECT * over-fetch, CTE re-scans, and MERGE target rewrites — live in the 23-rule engine today.

BigQuery product

On Snowflake

Waitlist

Coming soon: micro-partition prune misses, wrap-the-column filters, SELECT * over-fetch, and warehouse-credit waste from queries that never pruned.

Snowflake product

On Databricks

Waitlist

Coming soon: Delta full scans, missing partition filters, SELECT * on wide tables, and Photon shuffle-heavy jobs from system.query.history.

Databricks product
How it works

Stop paying for bytes nobody needed

  1. 01

    Scan query history

    30 to 90 days of jobs (by plan), plans, and bytes or slot time — never table contents.

  2. 02

    Match a documented rule

    Each finding names the table, the missed predicate, and a sample query.

  3. 03

    Fix the SQL or the layout

    You apply the change, or the Cognis AI DBA opens a pull request. The next scan is the scoreboard.

Blog

More from the blog

Databricks system table catalog path into ranked cost findings

Diagnosing Databricks costs with system tables

system.query.history and system.compute.warehouses rank full scans, idle warehouses, and shuffle-heavy jobs. SlashPeak never needs SELECT on your tables.

13 min read
Shuffle exchange where every map node sends rows to every reduce node

Databricks Photon shuffle costs and join key layout

Photon speeds up a shuffle join; it does not shrink it. Adaptive execution helps at the edges, but an unclustered join key still burns DBUs every run.

10 min read
Z-order curve versus liquid clustering over Delta files

Z-ordering vs. liquid clustering in Databricks

ZORDER and liquid clustering write min/max stats so Photon can skip files. The wrong columns, or a stale table, means it still reads everything.

10 min read
Databricks SQL warehouse DBU chart with idle auto-stop tail

Databricks SQL warehouses: auto-stop and idle DBUs

A serverless warehouse with a 10-minute auto-stop bills 10 minutes after the last query. Oversizing is a separate leak, and system tables show both.

13 min read

Take control of your warehouse bill

Understand, reduce and govern your cloud data spend with SlashPeak — starting today.

Start for freeGet a demo
  • 30-day free trial
  • No credit card
  • Read-only access
  • Set up in minutes