I am planning to create analytics database with Postgres which is connected with Metabase.
As I want to use Metabase model feature, I wonder how the model and persisted data are stored in the database. Is model stored as JSON in configuration database(not analytics database)?
Also is this ideal/ feasible?
Copy production database (Postgres) into analytics database (periodically) without breaking Models created by Metabase? What sort of thing shall be checked? Or any alternative suggestion is welcome!!!
Models are essentially queries and Model Persistence is a way to "cache" the results of those queries to make models load faster.
As long as the tables that your models build upon keep the same schema, the queries for the models will continue working normally, so you can insert/update/delete into that table as much as you want (using something like an ETL tool that takes data from your production DB and puts it into the analytics DB).
The model results would be stored in a separate schema inside the analytics DB in your case.
Let me know if that makes sense or if I missed anything
