Java programmers: if you are curious about how an SQL database works, take a look at H2's source (https://github.com/h2database/h2database). You can even try to fix an outstanding issue. It is illuminating - and much easier than you would have thought.
Simplicity is the big thing here; I can configure Spring to launch an in-memory H2 database and not have to worry about state, clearing and reseeding, or any of that stuff. The integration tests can just run.
I've been so happy with it that I'm pushing a customer to use H2 in production. I would normally use Postgres, but my customer doesn't really have an IT staff, and an embedded database will be a lot easier for them to keep up and running.
Just understand that any use that even vaguely resembles production will usually get pushback from people who think it is only a dev-level in-memory-only toy that causes constant slowdowns. This will be true even if it is only used for a few hundred rows worth of data.
Though I’d probably use SQLite for this use case