Horizon Context: Metadata connectors and external lineage in Snowflake

Use cases

Imagine you need to track data lineage coming through BI tools, external databases, and dbt models before landing in Snowflake. With Horizon Context, you can:

• Connect Horizon Catalog to 18 external sources, including Power BI, Tableau, dbt, Databricks, and BigQuery.
• Trace column-level lineage across different platforms.
• Manage external metadata, such as descriptions and tags, natively in Horizon Catalog.
• Delegate metadata curation using the APPLY COMMENT privilege without needing admin roles.
• Automatically propagate column comments through lineage.

How to use

To manage metadata access without granting data modification privileges, use the APPLY COMMENT privilege.

GRANT APPLY COMMENT ON TABLE my_catalog.my_schema.my_table TO ROLE my_data_role;

What happens in this code:

1. Grants the specific permission to update comments.
2. Allows data stewards to curate metadata without ownership privileges.

Limitations and requirements

• The feature is in Public Preview.
• Available in all AWS and Azure commercial regions.
• Not available in government regions or China.
• Lineage through temporary tables is generally available.

Using it in a pipeline

You connect tools like dbt and Databricks to Horizon Catalog so that data lineage is preserved and visible end-to-end, simplifying governance from ingestion to final reports.