Hi @abhijit007
For a Redshift --> Databricks migration, Lakebridge is designed to automate the code and metadata side of the migration and help you validate results on Databricks. Lakebridge does not copy data out of Redshift itself. Data movement is typically handled via Databricks Lakeflow/native connectors/cloud data-migration tools, with Lakebridge used to profile the estate and then validate the data on Databricks once it has landed. Lakebridge can scan your Redshift SQL and objects, estimate migration effort, and convert a large portion of SQL and DDL into Databricks SQL, surfacing what needs manual review. Lakebridge also focuses on the SQL and ETL logic inside pipelines. Orchestrators (e.g., Airflow, Step Functions, schedulers) are typically re-implemented or integrated with Databricks Lakeflow Jobs as part of the overall migration plan.
Lakebridge supports most of the technical migration lifecycle such as discovery & assessment which involves Profiling + Analyzer to inventory objects, classify complexity, and estimate effort and.... conversion which involves automated SQL/ETL conversion using a combination of deterministic converters and LLM-based translation for more complex patterns. It will also help with reconciling schemas and data (row counts, aggregates, column checks) between Redshift and Databricks to build confidence before the cut-over. However, it does not replace overall project management, cut-over planning, or data-movement plumbing, but plugs into those processes.
From a licensing and cost perspective, Lakebridge is a free Databricks migration tool. There is no license cost or consumption-based fee for using it. Parts of it (like Profiler and Reconcile) are open source. The Analyzer and converters are free but not open-sourced.
For Redshift to Databricks, we generally recommend:
- Use the Redshift to Databricks Migration Guide to structure the project into clear phases (discovery, assessment, design, migration, validation).
- Run Lakebridge to profile and assess the Redshift estate, then use its conversion capabilities to translate as much SQL/ETL as possible and highlight what needs manual refactoring.
- Use your preferred data-movement tooling to land data in the lakehouse, then use Lakebridge reconciliation (and optionally partner tools like Datafold) to verify that the migrated workloads are correct before cut-over.
Here are some references that may help.
Hope this helps.
If this answer resolves your question, could you mark it as โAccept as Solutionโ? That helps other users quickly find the correct fix.
Regards,
Ashwin | Delivery Solution Architect @ Databricks
Helping you build and scale the Data Intelligence Platform.
***Opinions are my own***