Skip to content
DataSquares
Start free

DataSquares Warehouse

A dedicated home for analytics data

Move analytical workloads into managed columnar datasets. Refresh from your sources on demand or on a schedule, and keep production focused on your applications.

Start freeBook a demo
Read the Warehouse documentation
  • Import a table or a saved query, then refresh on demand or on a schedule
  • Atomic refreshes: readers see the old copy or the new one, never a mix
  • Each workspace isolated by its own credential, enforced by the database engine

The Warehouse is an add-on on every plan, in three sizes. A company adds it on the Billing page in the app. See plans

A refresh your readers never notice

A refresh builds a staging copy beside the live one and swaps it in as one operation, so readers see the old copy or the new one, never a mix. Scheduled runs are recorded with their duration, rows and size.

What you can do with Warehouse

  • Built for analytical scans

    Columnar storage built for the scans dashboards run. Analysts read the copy, and your operational database keeps serving your application.

  • One more data source

    A dataset registers as a source. Build models, dashboards and saved queries on it with nothing new to learn and no separate API.

  • Atomic refresh

    A refresh builds a staging copy and swaps it in one operation. If it fails, the live copy keeps serving and the dataset is marked with an error.

  • Schedules that fire

    Hourly, daily, weekly or a custom cron with a timezone. When a scheduled refresh fails, the dataset’s owner hears once, by email and by webhook.

  • Isolation by the engine

    Each workspace reads the Warehouse through its own credential, granted only its own database. A query that reaches for another workspace’s data is refused by the database itself.

  • Lineage and safe deletes

    Each dataset shows the pipelines that feed it and the assets that read it. A dataset that anything still references can’t be deleted until it’s repointed.
Atomic
The old copy or the new one, never a mix
Isolated
A credential per workspace, enforced by the engine
Traced
Upstream pipelines and downstream assets on every dataset

Inside Warehouse

An error that still serves

If a refresh fails, because the source is unreachable or the query no longer runs, the swap never happens. The dataset keeps answering from its previous copy, and the error badge tells you freshness is what you’ve lost.

  • Refreshes don’t queue or run in parallel
  • After downtime, a schedule fires once for the latest missed run
  • Refresh history records duration, rows and size for scheduled runs

How it works

  1. Step 1: Choose the data

    A source table or a saved query

  2. Step 2: Import it

    Name it and set an optional schedule

  3. Step 3: Build on it

    Models, dashboards and the SQL editor

  4. Step 4: Keep it fresh

    Refresh now, or on a schedule

Questions, answered

Something else? Ask our team or search the documentation.

Do we need the Warehouse to use DataSquares BI?
No. BI queries your databases and warehouses directly through Connect. The Warehouse is for when you want a managed copy that takes analytical load off a source.
How large can a dataset be, and how does refresh work?
An import loads the whole result in one pass, within a configured row ceiling. Above the ceiling the import is refused, never silently truncated. Every refresh is a full re-import.
Who can import and delete datasets?
There is no separate Warehouse role today: any workspace member can import, refresh and delete a dataset. API keys need dataset scopes to read or change datasets.

Works well with Warehouse

All products
IntegrateDataSquares PipelinesMove and shape data between your systemsIntegrateDataSquares ConnectConnect the systems you already runGovernDataSquares CatalogA shared catalog for trusted data

See Warehouse on your own data.

Start a free trial in minutes, or book a walkthrough with our team.

Start freeBook a demo