🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

sharded metal go brrrrrr. this is a single database doing 130GB/s of filesytem throughput. that's ~7 million write IOPs.

Cruz Sam Lambert Hey Cruz, let’s break down why this filesystem throughput is a big deal for database performance—because, wow, these numbers are wild. A single database hitting 130GB/s with ~7 million write IOPS (input/output operations per second) is straight-up insane. To put this in

best hack night yet. absolutely electric, talented hackers, impressive agent demos next one to be even better~ hosted by AgentOps 🖇️ at PlanetScale HQ

The weather has been amazing lately so we decided to buy some E7 Campus APs and broadcast free and fast across South Park so everyone can work outside.

Need a refresher on vector database lingo? Our docs walk through embeddings, quantization, brute‑force vs IVF, HNSW, DiskANN, and SPANN - why each matters and where it breaks. planetscale.com/docs/vitess/ve…

burak emre George Fraser there are Vitess clusters that are 100PB in size doing over 50 million QPS. We have achieved linear scalability planetscale.com/blog/one-milli… so the constraining factor is compute availability. the largest Vitess cluster i know of has 70,000 servers in it.

Horizontal sharding offers several benefits: ✔️ Increased write throughput ✔️ Faster backups and restores ✔️ Faster schema changes ✔️ Failure isolation ✔️ More predictable cost projection Our workflows make implementing sharding simpler than ever.

This is happening! José Valim will speak at the special SF Ruby x Elixir meetup at PlanetScale HQ already on Wed, June 18. José is not only the author of Elixir language, a member of core Ruby and Rails teams in the past, and an inspiring speaker–he's now the founder of

Tomer Moran we should blog about this but basically we have no single points of failure and we truly share nothing. everything is multi region by default and our operator does over 100 million failovers a year. essentially we have an extreme tolerance to failure.

Failover is the foundation of high availability. At PlanetScale we perform over 100k each month, 99%+ of them on purpose. It's used for version upgrades, node TTLs, and unexpected failures. So what is a failover?

Been using PS from day one, and it’s been infinitely reliable.

Trying something a little bit different tomorrow 👀

👉 PlanetScale

Your primary key matters more than you think. In MySQL, the PK determines how you data is arranged on disk, meaning it has an impact on both disk usage and query performance. Check out some examples ⬇️

Thanks PlanetScale

Let’s go!!!

Just having the best night right now with José Valim and Ruby and Elixir friends at PlanetScale 💙

🤖 PlanetScale with Prisma ORM is a popular database choice among AI companies with heavy workloads. To ensure optimal performance for these situations, we added new attributes to the Prisma schema, enabling devs to specify shard keys on Prisma models! prisma.io/docs/orm/overv…