Postgres at scale — with copy-on-write branching, data masking, separated storage & compute , 100% Postgres and your own cloud
Joined November 2020
- Tweets 1,395
- Following 418
- Followers 4,069
- Likes 1,210
330 Photos and videos
Over the past couple of months we sped up provisioning, branching, and waking up Xata Postgres databases.
The article (and video demo) shows what that feels like: a database sitting scaled to zero with the bill flat, waking the instant you run a query, then branching into an isolated copy you can hand an agent without touching the source.
Once branches are this fast, a Postgres per PR, per CI job, per agent run, per psql session all become practical.
Each one is a full isolated database with copy-on-write data, awake only for the minutes you use it.
1
1
1
383
Xata 🦋 retweeted
Jun 11
1000 Postgres branches for $1
@xata Postgres databases are now provisioned, branched, and woken up from scale-to-zero dramatically faster.
Fast, practically free copy-on-write branches mean you can spin one up per PR, CI build, agent run, or psql session.
1
5
16
1,485
Xata 🦋 retweeted
Jun 11
Over the past couple of months, we’ve sped up significantly the provisioning, branching, and waking-up of @xata Postgres databases. Branch creation times are down from 25 seconds to ~2 seconds on average.
This results in a much smoother experience for you and your agents. Here is a 3 minutes demo.
2
8
24
2,102
This was made possible with two new features of the Xata platform: the new Xatastor storage engine and warm compute pools.
Xatastor separates compute from storage and branches are copy-on-write: they share the parent's data and stores only the changes to the parent.
Warm pools keep Postgres clusters ready, so a branch is ready to use in about a second.
1
577
1000 pure vanilla Postgres databases for $1. No fork, no proprietary APIs, every extension already works. You can get started for free at xata.io.
The full write-up: xata.io/blog/a-thousand-post…
1
118
Today, @monicasarbu, Xata CEO, joined the Berlin Buzzwords panel "The Agent Era: How AI Agents Are Reshaping Data Platforms".
The premise: AI agents have quietly become some of the most demanding users of modern data platforms, and most weren't built with them in mind.
She was joined by speakers from Snowflake, Elastic, and MotherDuck on what broke and what had to be rebuilt.
That gap, agents spinning up databases faster than any provisioning model can keep up, is the reason Xata exists.
1
2
155
Full writeup with the rest of what changed: display density, full-screen results, EXPLAIN plans, JSON and CSV export.
xata.io/blog/query-editor-v2…
145
We rebuilt the metrics view for every Postgres branch on @xata. Out of a central observability vendor, into a per-cell @VictoriaMetrics stack.
Six weeks of work, no user-visible downtime in the console. Here's what we learned. 🧵
1
3
9
1,049
Once the new store had a real retention window, we deleted the vendor client, the routing, the override header, the flag, the config.
Full write-up: xata.io/blog/how-we-rebuilt-…
1
128
We open-sourced a Next.js Postgres starter with 6 Claude Code skills.
Each skill wraps the Xata CLI for one workflow: branch, migrate, complete, rollback, clone-with-anonymization, setup.
Tutorial: xata.io/blog/building-produc…
1
2
9
527
254