News for hackers. Stuff that matters.

A comment on Xata: Postgres at scale, with copy-on-write branching and anonymization.

Re:Xata: Postgres at scale, with copy-on-write branching and... by beeman on Saturday May 17, 2025 @11:25PM (#44018800)
One of my mentees worked with Xata in their app using Prisma and it wasn't great. The need for a second db for the 'shadow' db that prisma needs, and they would throw Out of Memory errors frequently without any significant usage.

The app now moved to Prisma's postgres hosting and it works like a charm, only thing that changed is the db.

[ Parent | on HN ]
Re:Xata: Postgres at scale, with copy-on-write branching and... by nop_slide on Sunday May 18, 2025 @12:05AM (#44018934)
Oh yeah that prisma that decided doing joins in memory was a good idea? Yeah def would trust them to run a db hosting platform too

/s

https://github.com/prisma/prisma/discussions/19748

[ Parent | on HN ]
Re:Xata: Postgres at scale, with copy-on-write branching and... by sgarland on Sunday May 18, 2025 @11:13AM (#44021968)
I'm so glad I'm not the only person pointing this out every time they're brought up. I don't think I've ever felt the need to name and shame a company before, but this was just an egregiously horrible decision that they tried to smooth over and explain away. I will never have any faith in them whatsoever, due solely to that.
[ Parent | on HN ]
Re:Xata: Postgres at scale, with copy-on-write branching and... by tudorg on Sunday May 18, 2025 @02:31AM (#44019427)
[Author here] I'm sorry for the bad experience with Xata. If it's still an issue I'd be glad to have a look.

Just to make sure, I suppose this is on the Xata Lite free tier, not on the Postgres at scale platform that this blog post talks about.

[ Parent | on HN ]