How CARTO uses Object Storage (blob buckets)

Understand how CARTO uses object storage (blob buckets) across cloud providers

Overview

CARTO Self-Hosted uses object storage (also called blob storage or bucket storage) to handle data and assets used by the platform: imported datasets, generated map thumbnails, customization assets, and optionally data exported from your data warehouse.

You provide and manage these buckets in your own cloud account. CARTO connects to them using the credentials and configuration you supply. Supported storage providers:

Provider	Service name
Google Cloud	Cloud Storage (GCS)
Amazon Web Services	Simple Storage Service (S3)
Microsoft Azure	Azure Blob Storage

What CARTO stores in object storage

Every CARTO Self-Hosted installation uses two mandatory buckets and one optional bucket. The names shown in the Admin Console (Kots) and in customizations.yaml (Helm) are different. Use the tab for your deployment method:

☸️ Kots (Admin Console)

Bucket	Required	Purpose	Example contents
Temp Bucket	Yes	Upload and import datasets into CARTO	.csv, .geojson, .zip
Assets Bucket	Yes	Generated map thumbnails and customization assets (logos, markers, etc.)	.png
Export Bucket	Optional	Export data from your data warehouse — create only if you use data export features	.csv, .json, .parquet

🛞 Helm (customizations.yaml)

Objects in the assets / thumbnails bucket (such as map thumbnails) are publicly readable by default, so features like custom branding and map markers render correctly. The import/temp and export buckets are private.

Ownership and management

The buckets are part of your infrastructure: you create them, control access to them, and manage their lifecycle and cost. CARTO only reads from and writes to them using the access you grant. This keeps full ownership of the stored data with your organization.

Next steps

• Object Storage prerequisites — what to provision before deploying

• Configure your own buckets — step-by-step configuration guide