A catalog connects e6data to a metastore that describes your databases, tables, and columns. On AWS the most common path is AWS Glue over data in S3. Your S3 data is reached through the cross-account IAM role you configured earlier; the catalog supplies the table metadata.
This page covers the AWS Glue path for a Serverless deployment. For other catalog types and full catalog management, see Catalogs.
Before you start
Make sure you have configured AWS access: the cross-account IAM role exists, has S3 and Glue read permissions, and its ARN is registered in Settings → Storage Credentials.
Connect an AWS Glue catalog
- In the Compute Plane, open the Catalogs section and select Create Catalog.
- Enter a Catalog Name - unique within the workspace; lowercase letters, numbers, and hyphens recommended (for example,
aws-glue-catalog).
- Set Catalog Type to AWS Glue.
- (Optional) Enable Governance to apply fine-grained access control through Ranger, OPA, or Lake Formation. Use this if your organization enforces centralized data governance.
- Configure the connection:
- AWS Region - the region where your Glue catalog lives.
- Glue Database Filter - (optional) restrict to specific databases.
- (Recommended) Use shared Storage Credential — turn this on and select the Storage Credential you created earlier from the dropdown. This is the preferred method and reuses one keyless identity (the cross-account IAM role) across catalogs.
- Select Test Connection to verify connectivity and permissions.
- Choose the schemas to include by selecting the databases you need.
- Select Create Catalog. The catalog is now available for querying in your workspace.
Other catalog types
e6data also supports Hive Metastore, Databricks Unity, Apache Polaris, Iceberg REST, and Amazon S3 Tables. For Serverless deployments, each uses the same Cross-Account Access pattern (Target AWS Account ID + IAM Role ARN). See the per-catalog guides under Catalogs.
See also
Last modified on September 29, 2026
