Our last #P99CONF breakouts feature speakers who challenge the status quo. In 15 minutes, @tursodatabase's @Peristocles1, @almoggavra, and @UnikraftCloud's @felipehuici dive into topics like SQLite, LSM trees, and the cloud. Join them here > ow.ly/36yf51008Tu
#ScyllaDB
Proven strategies are on the way in these 11:20 am PST #P99CONF breakouts.
- Benjamin Cane: Mo Requests, Mo Problems
- @jaredhulbert: Memory is Slow, Disk is Fast
- @Evanfchan: 14 GB/sec io_uring + tokio+ Arrow + Rust Pipeline
Join us here: ow.ly/fx9A510058m
#ScyllaDB
Our next #P99CONF keynote features @TigerBeetleDB's @georg_kreuzmayr. In 15 minutes, Georg shares what it takes to apply TigerStyle to cloud I/O from first principles, and what you gain when every layer of the stack is something you own. ow.ly/JN1O51004jb
#ScyllaDB
(1/2) In 15 minutes, speakers from @PaloAltoNtwks, @Columbia, and @ScyllaDB take the #P99CONF stage.
- Prashant Pathak & Rama Teja Repaka: Eliminating Query Latency at Cloud Scale
- @vjabrayilov7: How We Hit Bare-Metal Tail Latency Inside Cloud VMs: Machnet
More technical #P99CONF breakouts are starting at 10:05 am PST.
- Ershaad Ahamed Basheer: #eBPF for Metrics Collection
- @forked_franz: Capturing Lightning in a Bottle
- Piotr Korcz & Kamil Dalidowicz: Integrating QUIC into Seastar
Join us > ow.ly/ceHV51003GL
#ScyllaDB
P99CONF retweeted
SQLite allows one writer at a time. Turso’s MVCC engine with BEGIN CONCURRENT removes that limit, and v0.8 is where the performance shows.
p99.9 transaction latency at 32 connections:
SQLite: 1.2 s
Turso: 2.4 ms
@Peristocles1, walks through the architecture behind these numbers at @P99CONF today, 11:45 am PT / 8:45 pm CEST.
P99CONF retweeted
memory/cache hierarchy from the fastest to the slowest @turbopuffer considers NVMe as part of the cache, which is becoming quite common for cloud native object storage solutions. #P99CONF @P99CONF
Every developer should keep these in mind.
Day 2's first #P99CONF breakouts start in 15 minutes!
- @lemire & Francisco Geiman Thiesen: Billion-Query-Per-Second Lookup
- @natevanben: Searching 100B Vectors on Object Storage
- Ivan Goncharov: Tracegrams
You can join us for free here: ow.ly/xI4V51003nO
#ScyllaDB
P99CONF retweeted
Speaking on this topic at @P99CONF in a couple hours!
Memory is Slow, Disk is Fast - part 3
Can an on-disk data structure with insert(), delete() be faster than an in-RAM binary tree?
New Blog Post up.
bitflux.ai/blog/memory-is-sl…
P99CONF retweeted
Interesting, around 50% of attendees manage max 10TB of data (no DB specified) insight from #P99CONF.
Binary heaps are cache-inefficient and cause a cache miss on nearly every pop. In 20 minutes, @curious_coding dives into SimdQuickHeap and how it results in a priority queue that is up to 10x faster than binary and d-ary heaps. ow.ly/bcB951003al
#ScyllaDB #P99CONF
P99CONF retweeted
Turso concurrent writes talk by @Peristocles1 at @P99CONF today! Dont miss it
Turso is at @P99CONF this week 👇 (online, free, Oct 21–22):
@Peristocles1 will cover breaking SQLite’s single-writer bottleneck with MVCC
Thanks to @ScyllaDB for having us!
We hope you're ready for an action packed Day 2 of #P99CONF! Don't miss keynotes from @DorLaor, @curious_coding, @georg_kreuzmayr, and @reconditerose. We also have 15 breakouts today, so there is something for everyone! Join us for free here > ow.ly/cL6e51002vI
#ScyllaDB
P99CONF retweeted
It's crazy what all it takes to design a time series database capable of processing millions of transactions per second plus bound to the time. Thanks for the great talk @hagen1778 and @ScyllaDB @P99CONF #P99CONF
We wrap up Day 1 of #P99CONF with a jam-packed keynote from @fasterthanlime in 10 minutes. Come learn about Stax, a profiler for macOS that gives engineers a holistic view of performance and is friendly to agents & remote work alike. Join Amos: ow.ly/Cpwv50ZZLpE
#ScyllaDB





































