Some literate programming on a small crate I've written Read more ›
NOTE: This is a pre-release feature. The Fil-C 0.679 release does not ship with this feature. To test this feature, you need to build from source. Read more ›
As a follow-up to last week's article around Firefox leveraging zlib-rs and some nice upstream improvements to this Rust-based Zlib implementation, the zlib-rs 0.6.4 release is now available to ship all of these latest enhancements... Read more ›
ik_llama.cpp fork with experimental NUMA per-node mirroring of model weights and KV cache. This is an attempt to maximize CPU inference performance on multi-socket systems. - mikechambers84/ik_llam... Read more ›
Proper TCP socket splicing reduces the load on userspace processes and enables more efficient data forwarding. We realized that Linux Kernel's SOCKMAP infrastructure can be reused for this purpose. Read more ›
I'm writing an article series on Free Threading internals to learn more about Free Threading, explain how it works, and explain how it solved the "remove the GIL" issue where previous attempts failed. replaces the unique Global Interpreter Lock (GIL … Read more ›
Executive Summary: To eliminate race conditions in a high-concurrency ticketing system, I implemented PostgreSQL's FOR UPDATE clause for row-level database locking alongside Go worker channels for in-memory queue serialization. This approach completely prevents inventory over-selling by guaranteeing singular data mutation execution even under flash-sale load. If you have ever built an application that handles live event ticketing, flash sales, or limited-inventory drops, you know the dread of... Read more ›
Every time a new Java release drops, the question surfaces again in Slack channels and pull-request threads alike: “Are we still on G1? Should we switch to ZGC?” With Java 26 landing in March 2026, that conversation got a meaningful update. Two JEPs directly touched garbage collection — one gave G1 a solid throughput boost and the other … Read more ›
I've been experimenting with USTP-Secure, a reliable UDP transport I've been building. The latest version adds: Optional congestion control Retry tokens to reduce amplification risk Optional cleartext + HMAC mode (integrity without encryption) Authenticated ACK/NACK packets One design choice I'm still thinking about is congestion control for an intentionally unordered reliable transport. Unlike TCP, USTPS doesn't enforce ordered delivery at the transport layer. Missing packets can be retransm... Read more ›
A visual and written explanation of Distributed Compaction (RFC-0025 in SlateDb) and discussion of future work Read more ›
Oversteer manages steering wheels on Linux using the features provided by the loaded modules. Features: Change rotation range. Change emulation/working modes. Combine accelerator/brakes pedals for games that use just one axis. Change autocentering force strength. Change force feedback gain. Device configuration profiles. Overlay window to display/configure range. Use wheel buttons to configure range. Hardware performance testing. Combine accelerator/clutch pedals. Useful for flight simulators... Read more ›
ZFS and Ceph are both powerful open-source storage technologies, but they address fundamentally different challenges. This article explores the trade-offs between distributed storage and standalone storage architectures, helping infrastructure teams determine when Ceph’s scalability is necessary and when ZFS offers a simpler, faster, and cost-effective solution. The post appeared first on . Read more ›
The seL4 organisation on GitHub uses As an experiment, I have taken the various manifest files across the org, and constructed a graph based on how frequently each pair of repositories is mentioned in a manifest together. See below: [This may render badly when syndicated outside of my blog; and also on small screens. And probably large screens. I’ve attempted to make sure there’s a – on my site with JS enabled, if you hover over a node, it should highlight connected nodes.] The colouring of t... Read more ›
Conflict-free Replicated Data Types (CRDTs) ensure Strong Eventual Consistency without coordination, but typically assume benign participants and rely on validation or exclusion to handle Byzantine behavior. We address this problem through deterministic state reconstruction: rather than deciding which updates are admissible, all accepted updates are incorporated, while only a subset contributes to the reconstructed state. We instantiate this app... Read more ›
This report provides an in-depth analysis of the major x86 hypervisor and emulator projects that define modern virtualization infrastructure. The landscape spans from bare-metal... Read more ›
The line-delimited JSON protocol the rad CLI uses to talk to a running radicle-node over a Unix domain socket. Read more ›
First, I want to tell you how exactly I got to this point and why I started researching different options for handling asynchronous I/O on Linux… Last year, my students and I built a reverse proxy server called TinyGate. It was super simple, worker-based, and it basically worked well. Of course, I didn’t expect it to be very fast, but it was an educational project, and since we’d made a real, kind of production-ready tool, I was really proud of it. But my students weren’t as happy as I was - ... Read more ›