Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Migrating an existing application to Citus sometimes requires adjusting the schema and queries for optimal performance. Citus extends PostgreSQL with distributed functionality, but row-based sharding isn't a drop-in replacement that scales out all workloads. A performant Citus cluster involves thinking about the data model, tooling, and the choice of SQL features to use.
There's another mode of operation in Citus called schema-based sharding. While row-based sharding results in the best performance and hardware efficiency, consider schema-based sharding if you need a more drop-in approach.
Step 1: Optimize the schema
The first steps are to optimize the existing database schema so that it can work efficiently across multiple computers.
Step 2: Update application queries
Next, update your application code and queries to deal with the schema changes.
Step 3: Migrate production data
After you test the changes in a development environment, the last step is to migrate production data to a Citus cluster and switch over the production app. We have techniques to minimize downtime for this step.