PlanetScale

Kicking off PG Day the right way. We're taking over Golden Goat Coffee on Nov 4 with the Golden Elephant, our drink of the day, free from 9 AM to noon. Grab a ticket:

Postgres is a go-to choice for a lot of teams now, but it hasn't always been. For almost 40 years, it fought off competition and near-death. What was its defining moment on the way to inevitability? Let's take a look.

Did you know you can use VIM inside of Postgres?

Shard groups are the primitive for scaling Postgres with Neki. Here’s how they work:

With ON DELETE CASCADE, you're only one query away from a complete outage. Tom explains why…

Back in 2009, NoSQL took web development by storm. What happened since?

It's been one month since we released the Neki platform preview. Since then we've added $50 Metal nodes, improved Postgres compatibility, read-only replicas, Azure support, and more. Full list of updates and where to start:

Postgres went open source 30 years ago, and we're celebrating Nov 4 with a PGDay in SF. We'll be joined by Bruce Momjian, co-founder of open-source Postgres, Fabrizio Mello from our Postgres core team, and more. RSVP: luma.com/haxfnzo1

we're getting some engineers together for ramen at @planetscale.com's office. good food, drinks, and a chance to hang. luma.com/jvrl5pcb

PlanetScale is now available on Azure planetscale.com/blog/planets...

Smaller isn't always faster. What gets TOASTed in Postgres and when to keep pglz:

AWS's own docs say gp3 volumes can drop below 90% of provisioned IOPS 1% of the time. That's 86 hours a year, per volume. Spread a database across 768 of them and there's a 99.65% chance one is hurting production right now.

Join leaders from Laravel, PlanetScale, and WorkOS for a discussion on building and scaling their agent stacks. RSVP today.

TIN is the fastest text search for Postgres. Again.

Celebrate the end of SF Tech Week at Developers After Dark, an evening of technology, nostalgia, and after-hours fun at the Exploratorium. RSVP now.

Imagine a database that can reject work before it even starts, based on available resources, to maintain its own health. It's possible now, and its origins are ancient.

Moving keyword scoring to PlanetScale TIN took one Layers storefront search from ~264 ms to ~51 ms. A 5x improvement, all by using the right tool for the job. Read more: layers.engineering/how-planetsc...

"We wanted settings changes to be boring" @triggerdev.bsky.social on why they picked PlanetScale Metal for running Postgres: local NVMe so vacuums don't slow down prod, easy configuration, and outstanding support.

Recently, Shopify switched an integral part of their shopping cart system from Redis to MySQL. Is Redis dead?

Most teams put up with a database that has outgrown its home because migrations feel too risky. We regularly move terabyte and petabyte databases with no downtime, and reverse cutover if we need to.

We aimed for 1 million queries per second on sharded Postgres. Five shards got us there. Then we kept going: 512 shards, 1.22 PiB, 118.5 million QPS.

You can now run a zero downtime MySQL import straight from the CLI. Coming from RDS, Aurora, or Cloud SQL? Run the whole migration from your terminal or hand it off to your agent to do for you.

Nothing is as fast as not doing anything. How we build Neki around that idea:

What happens to your database if the data center goes down? Build it right and it just keeps working.

Ever wonder what happens to a query when your Postgres database is spread across 1000s of servers? Follow one through auth, the wire protocol, parsing, planning, and connection pooling, then back to your app with the results.

Suffering from success? When a huge tenant starts stressing your shards, Neki has the tools to scale up, isolate, and reshard without stress.

ARM or x86? Your choice has a big impact on Postgres performance. Here's what to know before you pick: planetscale.com/blog/when-to...

Join us for PlanetScale Movie Night on October 17. We'll be watching "You Can See Everything" at AMC Kabuki 8 in San Francisco. Tickets are limited, so make sure to RSVP: luma.com/qjw9i5a9

SELECT name FROM search_engines WHERE description ==> '"fast text search"'; name ------ TIN (1 row) Read the blog: planetscale.com/blog/searchi...

PlanetScale is at Rails World this week in Austin, Texas. Come by our booth for merch, a product demo, and to meet some of the team.

How do you make text search fast? Inverted indexes, tokenizers, BM25, segment merges, and more. Our latest article covers what it takes to build a search engine inside Postgres.

How do you build the fastest text search in Postgres? Ask the two engineers who did it. TIN TALK: live Q&A with Eric and Patrick on our Discord, Monday 11am PT. Join to ask your questions: pscale.link/community

A Postgres replica can finish replaying WAL and still not be safe to promote. If it doesn't have your logical replication slots, your change stream breaks on cutover. See how PlanetScale solves this:

Let's take a closer look at how Neki is built. Starting with real Postgres, we unpack the architecture one component at a time:

We wanted more from full text search in Postgres, so we built TIN. It gives you fuzzy matching, NEAR, wildcards, boolean grouping, and BM25 scoring. Plus, it's really fast. See it in action in our latest video:

It was a packed house last week for the SF Database Meetup. Our CTO @nvw.io took the stage to break down sharding, why we built Neki, and why it's much more than just sharded Postgres. Full talk on YouTube:

Join us for the @changelog.com IRL at PlanetScale HQ. We'll have three special guests joining us to discuss background agents, developer tools, and more. Happening in San Francisco this Thursday, September 17th. RSVP:

S3 doesn't work the way you think it does. In this video, we uncover how S3 handles millions of requests per second on basic hard drives.

118,538,803 queries per second. 1.22 PiB of data. 512 shards. Powered by Neki. planetscale.com/blog/118-mil...

Neki, horizontal sharding for Postgres, is available today. Join us tonight for a first look at Neki with talks on Neki's architecture, why you should shard, the lifecycle of a query, and data movement between shards.