PlanetScale
@planetscale.com
The database platform built for scale. Postgres and Vitess/MySQL. https://planetscale.com Discord: http://discord.gg/pDUGAAFEJx Status: http://planetscalestatus.com
Shard groups are the primitive for scaling Postgres with Neki. Here’s how they work:
Shard design doesn't start from your tables — PlanetScale
Shard group design starts from your queries, not your tables. Find the key that your most important queries and transactions share, and put every table that carries that key in one group.
planetscale.com
With ON DELETE CASCADE, you're only one query away from a complete outage. Tom explains why…
I don't trust ON DELETE CASCADE — PlanetScale
A one-line convenience in your database schema can become an outage. There are better ways to delete data that won't risk locking up your database.
planetscale.com
Back in 2009, NoSQL took web development by storm. What happened since?
What happened to NoSQL?
Check out https://planetscale.com for more Postgres and MySQL conte...
youtube.com
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:
Neki update: $50 Metal, read-only replicas, Azure, and more — PlanetScale
Postgres compatibility, read-only replicas, smaller Metal instances, Azure, and more have landed for Neki since platform preview.
planetscale.com
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
Postgres Day SF Meetup · Luma
Join us on November 4th at PlanetScale’s San Francisco office as we celebrate 30 years of Postgres. We’ll be joined by Bruce Momjian, co-founder of open source…
luma.com
huge_pages is a huge deal for huge workloads. Available now for Postgres and Neki databases.
we're getting some engineers together for ramen at @planetscale.com's office. good food, drinks, and a chance to hang. luma.com/jvrl5pcb
🍜 Ramen pop-up // PlanetScale x Depot · Luma
🍜 Ramen pop-up // PlanetScale x Depot Join PlanetScale and Depot for ramen and drinks on November 5. The Depot team is in SF, and we’re teaming up to turn the…
luma.com
Smaller isn't always faster. What gets TOASTed in Postgres and when to keep pglz:
What gets TOASTed in Postgres — PlanetScale
Postgres 19 switches TOAST compression from pglz to LZ4. LZ4 is faster, but it isn't always smaller, and some reads get slower with it.
planetscale.com
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.
The Real Failure Rate of EBS — PlanetScale
Our experience running AWS EBS at scale for critical workloads
planetscale.com
Join leaders from Laravel, PlanetScale, and WorkOS for a discussion on building and scaling their agent stacks. RSVP today.
Founders Panel: The Agentic Stack · Luma
Founders & leaders from Laravel, PlanetScale, and WorkOS talk building and scaling on their agent stacks. They’ll discuss: Where agents are helping Risks…
luma.com
TIN is the fastest text search for Postgres. Again.
TIN v1.0.6: 2-3x faster, with support for stemming — PlanetScale
TIN v1.0.6 supports stemming in 18 languages and is 2-3x as fast as the GA version
planetscale.com
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.
Developers after Dark · Luma
Celebrate the end of #SFTechWeek at Developers After Dark, an evening of technology, nostalgia, and a little after-hours fun at the Exploratorium. We're…
luma.com
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.
What today's software owes to Bronze Age clocks — PlanetScale
Admission controls in software determine what does or doesn't happen next, and the most common implementation has ancient roots.
planetscale.com
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...
How PlanetScale TIN Became the Foundation for Layers Search
Quick note first. If you’ve been following TIN, you’ve seen the takes: it’s fast, “is it fast?”, dense-term elision is a mistake, etc. We…
layers.engineering
Everything's faster when your database is close to your app. Today we're launching dedicated read replicas for Postgres. Run replicas across the globe, near both your apps and your end-users.
"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.
Upgrading a 56TB database that's growing at 250GB per day with (nearly) no downtime | Trigger.dev
We had to upgrade our 56TB Aurora Postgres cluster by 31 July without taking Trigger.dev down. Instead of moving the database, we made the problem smaller: new runs and the control plane went to PlanetScale, and the old runs stayed behind.
trigger.dev
Recently, Shopify switched an integral part of their shopping cart system from Redis to MySQL. Is Redis dead?
is Redis cooked?
Check out https://planetscale.com for more Postgres and MySQL conte...
youtube.com
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.
Zero downtime migrations at petabyte scale — PlanetScale
Data migrations are a critical part of the database lifecycle, and are sometimes necessary for version upgrades, sharding, or moving to a new platform. In many cases, migrations are painful and error-prone. In this article, we walk through how migrations are performed at PlanetScale, and offer advice on how to improve the migration experience.
planetscale.com
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.
118 million queries per second on Neki — PlanetScale
We ran a massive, sharded Postgres database at 118.5 million queries per second, with 200k queries per second on each shard across 512 shards.
planetscale.com
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.
Import your MySQL database with the pscale CLI — PlanetScale
You can now import an existing MySQL database into PlanetScale Vitess with no downtime from the pscale CLI. Connect your database to a production branch as an external keyspace, then move its tables into PlanetScale with a Vitess MoveTables workflow. Your database keeps serving traffic while the data copies and replicates, and you switch reads, then writes, when you're ready.pscale keyspace create-external mydb main mydb_source \
planetscale.com
Nothing is as fast as not doing anything. How we build Neki around that idea:
Designing Neki for performance — PlanetScale
How Neki only decodes the parts of a PostgreSQL result the router actually needs.
planetscale.com
What happens to your database if the data center goes down? Build it right and it just keeps working.
The data center caught FIRE (but my app is fine)
AWS regions and availability zones are how applications stay online...
youtube.com
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.
The lifecycle of a sharded Postgres query — PlanetScale
Follow a SQL query through the router, across four Postgres shards, and back.
planetscale.com
Suffering from success? When a huge tenant starts stressing your shards, Neki has the tools to scale up, isolate, and reshard without stress.
Handling hot shards — PlanetScale
What makes a good sharding strategy today might not work so well tomorrow. What happens when you have one tenant hitting your database orders of magnitude more than any other? Or product managers knocking down your door with new requirements? If you're already on Neki, fortunately these are issues we can solve.
planetscale.com
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...
When to choose x86-64 vs aarch64 — PlanetScale
ARM and x86 can give you different performance at the same vCPU count. Here's what differs, what to measure, and how to choose.
planetscale.com
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
PlanetScale Movie Night: You Can See Everything · Luma
Nathan Fielder made a documentary about Elizabeth Holmes, we got a bunch of tickets to watch it with you. Join us at AMC Kabuki 8 to watch You Can See…
luma.com
SELECT name FROM search_engines WHERE description ==> '"fast text search"'; name ------ TIN (1 row) Read the blog: planetscale.com/blog/searchi...
TIN Postgres search is faster, better, and cheaper — PlanetScale
A good Postgres search demo was doing the rounds online. Let's see what happens when we upgrade it to TIN.
planetscale.com
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.
