2. DID interning gave an almost 6x reduction in backfill write amp!!!!! write amp!!!!!! 36x -> 6.2x write amplification! hugeeee for running hubble on smaller VPSes with lower-throughput network block storage. i gotta write this one uppppp
fig (aka:[phil])
bonus backfill notes: 1. the order of repos from relay listRepos has a *significant* bias toward bigger-repos-first. this is not optimal for minimizing total backfill time!! 2. theres a huge backfill write amplification trick: interning DIDs. write amp!! not space! i gotta write this one up.
Jul 31, 2026 at 7:33 PM
Hubble first launch: 1500x cheaper getRepo for Bridgy Fed - microcosm
Bringing Bridgy's backfill back! (and unblocking a big Bluesky broadcast 👀)
VPS with the Best Price-to-Performance Ratio in the US | Contabo
High-end Cloud VPS hosting for a fair price. Featuring AMD CPUs, lots of RAM (from 8 GB), NVMe storage, and unlimited traffic, with root access. Spin up your cheap VPS in minutes!
Megakernel: Matching Apple Silicon Efficiency at 2x the Throughput on a RTX 3090
The first megakernel for hybrid DeltaNet/Attention LLMs. 413 tok/s at 1.87 tok/J on a 2020 RTX 3090, matching M5 Max efficiency at 1.8x throughput.
How Elon Musk's AI Empire In Memphis Became A Cautionary Tale
How three lines of configuration solved our gRPC scaling issues in Kubernetes
It all started with a question I asked our senior software engineer: “Forget the speed of communication. Is it really better for you to…
Small space big upgrade. https://t.co/6tUpFi1npD
Small space big upgrade. pic.twitter.com/6tUpFi1npD— LMD (Arc.) (@Layemie001) March 22, 2026
Paul Rony on Twitter / X
As models gets better and better, running them half-on-a-datacenter-half-on-a-laptop becomes less practical every day. That’s why we think that each user now needs its own machine, ready to do your best work.— Paul Rony (@Paul_Rony) August 21, 2026
A Full-Network Relay for $34 a Month | bryan newbold
This is an update to a Summer 2024 blog post. At the time, atproto relays required a cache of the full network on local disk to validate data structures. With the Sync v1.1 updates, relays don't need all that disk I/O. What impact does that have on hosting setup and operating costs? Turns out the d...
Saturation: How Your Software Will Fail at Scale - Lorin Hochstein | SSW 2026
OpenAI Chat GPT OSS 20b Open Source LLM Full Local Ai Review
"The end result of all of this is that we grew from 2k users to almost 150k, added a ton of heavy new functionality, and still managed to optimize and cut down costs from $.15 per active user per month to just $.03 or so." - @snarfed.org breaks down all the work he's done to optimize Bridgy Fed 1/
Bridging on a budget
blog.anew.socialfull-network space efficiency update: bad news: event though 84% of all repos are stale (no activity in 30 days), stale repos only make up 20% of the data in the network. moving them to cheap storage doesn't buy much. good news: the data in network is small?? like sub-1TB, minimal magic needed
Added a new device to my @tiles.run cluster. Welcome to the lineup, M5 Pro with the 32 GB/1 TB spec.
