Staging on both sides
Databricks loads through its S3 or Azure staging and MotherDuck through your S3 bucket. This is the pair where you configure storage twice.
Connect Databricks and MotherDuck with Polytomic to move tables between the lakehouse and DuckDB. Both platforms load through external cloud storage, so staging is configured on each side, and Polytomic runs the ETL and reverse ETL between them on whatever schedule suits your work.
Databricks loads through its S3 or Azure staging and MotherDuck through your S3 bucket. This is the pair where you configure storage twice.
Databricks and DuckDB both model nested data properly, so structs, arrays, and maps move across without being serialized into strings.
Reads and writes run as the identity you connect with, so Unity Catalog permissions apply exactly as they would to any other client.

Bec Hu
Head of Sales Ops, Chowbus
“The support is among the best I have ever encountered.”
Follow these simple steps to connect MotherDuck with Databricks in minutes
Connect MotherDuck
Connect Databricks
Create sync
Enable sync
Connect MotherDuck to Polytomic with an access token and your database name.
Connect Databricks with a SQL warehouse hostname and token, plus cloud storage if you are writing.
Sync data in either direction between MotherDuck and Databricks. Choose the source data and destination objects for the direction you need.
Map the fields, choose the sync schedule and behavior, then save and enable the sync.
Get answers to common questions about the integration process
Sync data in either direction between MotherDuck and Databricks.
Use both integrations' source and destination capabilities in one managed sync workflow.
Enterprise encryption and security protocols ensure your data remains protected.
Get up and running in minutes with our step-by-step integration guide.