Skip to Content
Get startedSet up warehouses for the studio

Set up warehouses for the studio

The install creates a warehouse and a compute pool for the Data Product Studio. This article walks you through what they are, how to confirm them, and when to tune them.

Applies to: Snowflake admin Detail level: Standard

Who this is for

You’ve installed the Data Product Studio and granted it access to source data. The install created the studio’s warehouse and compute pool automatically, and they’re already running. This article covers what they’re for and when you might need to change them.

Before you start

You’ll need:

You should have already read:

  • Install from Snowflake Marketplace
  • Set up roles for the studio

Steps

  1. Open Snowsight and switch to a role that can manage the studio’s warehouse and compute pool.

  2. Find the warehouse the install created. Go to Admin > Warehouses and look for DATA_STUDIO_WH. The studio uses this warehouse to read the application database and to deploy data products.

  3. Find the compute pool the install created. Go to Admin > Compute Pools and look for DATA_PRODUCT_STUDIO_APP_POOL. The app’s services run on this compute pool.

  4. Confirm the defaults:

    • DATA_STUDIO_WH: X-Small, with a 1-minute auto-suspend.
    • DATA_PRODUCT_STUDIO_APP_POOL: runs the app’s UI and services.

  5. Decide whether to keep the defaults or scale up. Most accounts run the studio comfortably on X-Small. Consider scaling up if entity generation feels slow, or if several team members are using the studio at the same time.

Screenshot: Snowsight Warehouses page showing DATA_STUDIO_WH at X-Small with 1-minute auto-suspend.

Screenshot: Snowsight Compute Pools page showing DATA_PRODUCT_STUDIO_APP_POOL.

What happens next

The studio is ready for your data team to use. Send them to Data Product Studio Home and have them define their first data product. See Create your first data products.

Troubleshooting

DATA_STUDIO_WH doesn’t appear in Snowsight

Likely cause: the install hasn’t completed, or your current role doesn’t have visibility into the application’s warehouse.

Resolution: confirm the app is installed by going to Data Products > Apps. If the app is listed but the warehouse isn’t visible, switch to ACCOUNTADMIN or to a role with MONITOR privilege on the application.

Before contacting support: capture your current role, the install time, and a screenshot of the warehouse list.

Entity generation or metric generation is slow

Likely cause: DATA_STUDIO_WH is sized too small for the workload, or several jobs are queued on the same warehouse.

Resolution: check the warehouse load chart in Snowsight under Admin > Warehouses > DATA_STUDIO_WH. If load is consistently high, alter the warehouse to a larger size:

ALTER WAREHOUSE DATA_STUDIO_WH SET WAREHOUSE_SIZE = 'SMALL';

You can return it to X-Small later if usage drops.

Before contacting support: capture the warehouse load chart, the warehouse size at the time of the slowdown, and the timestamps of the affected runs.

The compute pool shows a “Resuming” or “Stopped” status when users open the app

Likely cause: the compute pool auto-suspended after a period of no use, and is restarting on the first new request.

Resolution: wait sixty seconds for the compute pool to resume, then reload the app. Repeated cold starts are expected for low-use accounts. If wait times are unacceptable, contact support to discuss compute pool sizing.

Before contacting support: capture the compute pool’s status history and the times users hit a cold start.

FAQs

Can the studio share its warehouse with other workloads?

The install creates a dedicated warehouse, DATA_STUDIO_WH, and the app uses it directly. Sharing this warehouse with unrelated workloads isn’t supported. If you need to isolate other workloads, create a separate warehouse for them rather than reusing DATA_STUDIO_WH.

Why does the install create a compute pool as well as a warehouse?

The compute pool runs the application’s services: the app itself, its UI, and its background streams and tasks. The warehouse runs the SQL workloads the app submits, mostly reads against the application database and deployments to your account. They have different jobs and are sized independently.

What happens if I drop DATA_STUDIO_WH?

The app stops working until the warehouse is recreated. Because the install grants the app CREATE WAREHOUSE, the app can recreate DATA_STUDIO_WH if it’s missing. The safer path is to alter the warehouse rather than drop it.

How big can DATA_STUDIO_WH get?

Start at X-Small. Most accounts don’t need more. If you see persistent queueing under load, step up to Small, then Medium. Beyond Medium is rare for the studio’s workload; if you’re considering it, raise it with support so we can look at the workload pattern with you.

  • Prerequisite: Set up roles for the studio
  • Sibling: Set up roles for the studio
  • Next step: Create your first data products
  • Troubleshooting: Resolve generation failures in the studio

Search keywords

configure warehouse, warehouse sizing, assign warehouse, native app warehouse, DATA_STUDIO_WH, DATA_PRODUCT_STUDIO_APP_POOL, compute pool, scale up warehouse, alter warehouse size, auto-suspend

Last updated on
Set up roles for the studioCreate your first data products