Is this similar to doing something like `rails db:schema:load` and then `rails db:seed`? i.e. set a brand new db to a certain schema + preload a set of fixtures into it?
The effect will be relatively similar -- you arrive at a new database with the proper schema/structure in place, but the difference is in execution. Doing this _within_ the database engine is likely 10x faster than loading the schema through a client library making a series of CREATE and ALTER statements (ala the schema:load rake task).