Docs Ordnary Web Services Ordnary R2
Data catalog
Querying a bucket's objects as Apache Iceberg tables from engines such as Spark, Trino and DuckDB. Beta.
The data catalog turns a bucket into an Apache Iceberg catalog: tables whose data files live in the bucket and whose metadata the catalog keeps, so query engines can read and write them with the Iceberg REST protocol. It is for analytics on data you already store in R2, without copying it into a warehouse.
Turn it on
On the create form under Advanced settings, or later under the bucket's Settings, R2 Data Catalog. Turning it on does not change existing objects; it adds the catalog endpoint for the bucket, which the settings page shows together with how to connect. The same access key authenticates engines.
Engines
- Apache Spark with the Iceberg REST catalog.
- Trino and other engines that speak Iceberg REST.
- DuckDB and PyIceberg for local analysis.