I've previously spent a fair bit of time tuning postgres for fast/less durable settings to speed up testing, and this puts all of that to shame when you need to start with any non-trivial schema/data and revert back to it.

I had been testing out some ideas with `docker commit` to save derived images which included a premade db, then reverting to that, but I don't think it's worth bothering with since I found dblab.

Haven't yet tried using it CI, and I suspect it might need a somewhat-custom-than-normal VM setup to use ZFS than you could easily do on most hosted CI runners, but it's on my list of things to investigate/setup eventually.

For more real-life feedback, welcome to the Database Lab Community Slack: https://slack.postgres.ai/

In one company, it grew into a cluster of a few servers for higher capacity and availability. This load balancing and management code is a quite more complex than the original single-server snapshotting script. Unfortunately, it's not open source. For several years, the single server was enough. Go for it. :-)