yesterday we processed 10B+ vector/document writes (~40TB) that's 100,000+ writes per second sustained
turbopuffer
Technology, Information and Internet
built on top of object storage, which means 10x-100x cheaper, usage-based pricing, and massive scalability
About us
- Website
-
https://turbopuffer.com
External link for turbopuffer
- Industry
- Technology, Information and Internet
- Company size
- 2-10 employees
- Type
- Privately Held
Employees at turbopuffer
Updates
-
changelog for february! 🐡 https://lnkd.in/eTVNvKs5
-
-
every time we re-run the landing page benchmarks, they improve slightly. (bars are off because these are simply website screenshots) https://turbopuffer.com/ by default, turbopuffer does consistent reads, but you can switch to eventual consistency for even faster queries.
-
-
A summary of what we shipped in January 🐡 https://lnkd.in/eTVNvKs5
-
-
changelog for november & december 📉🔢💫 https://lnkd.in/eTVNvKs5
-
-
what team tpuf shipped in September: 🔒 SOC 2 Type 2 📝 Blog post on Continuous Recall Measurement 🦣 8 MiB attribute value limit (up from 64 KiB) 🚁 Add copy_from_namespace to create a namespace by copying another namespace (50% discount relative to upserting from scratch) 🔳 Add uuid type (55% discount from string) and bool type https://lnkd.in/eTVNvKs5
-
Our first blog post is out "turbopuffer: fast search on object storage" https://lnkd.in/g3CGHhpK