Topics /
The most popular blog posts about systems & performance on Hacker News. 518 posts.
Firecracker needs KVM and therefore a Linux host, and the image tooling around it needs loop mounts and Linux userland. We rebuilt the stack on Apple's Virtualization.framework to boot the same OCI images as microVMs on macOS and on Linux.
Prototyping a Home NAS with RAID1 on FreeBSD on a virtual machine: system installation via SSH with bsdinstall, configuring ZFS mirror and working with datasets
Figuring out how it went missing.
How ParadeDB uses principles from search engines to optimize Postgres' Top K performance.
There is an odd bug in Windows related to Video CDs . To reproduce it: Insert a Video CD (if you don't have one, you can download this fi...
How long should a micro benchmark run? My rule of thumb is to tweak the input size until the benchmark takes about 300ms, for the following reasons:
Evidence-indexed documentation of TI-84 Plus OS 2.55MP ROM behavior and hardware interfaces.
Oxide's approach to emulation, simulation, and tiny versions of big things.
OpenSSL sucks. The BoringSSL and AWS-LC forks are Googled and Amazoned to death; they don't care about anyone but their own use cases. I can't remember ever having a good experience with software using GnuTLS. LibreSSL is incomplete... FOREWARD This post is about the experience of taking a leap of …
Meet Alice. Alice uses your web service. Alice, like most humans, measures her time in seconds and minutes. Alice says your service is slow. You tell Alice that the mean request to your service completes in 100ms, but Alice says that her mean wait time is 1s.
Git packfiles were designed for mmap and local disk, so pulling one object out of a bucket means guessing at a byte range. I wrote a new format instead.
Hamming Codes ECC RAM corrects errors that occur in memory before it gets to the CPU. The most common form of ECC is the Hamming Code [1] which when it has R redundant bits can correct single bit e…
Does a new algorithm that improves on Dijkstra'a classic one actually deliver a practical benefit for networking?
You are seeing this because the administrator of this website has set up Anubis to protect the server against the scourge of AI companies aggressively scraping websites. This can and does cause downtime for the websites, which makes their resources inaccessible for everyone.
How vLLM reaches 2.2k tokens per second per H200 for DeepSeek-style MoE serving with Wide-EP, async scheduling, dual-batch overlap, disaggregated serving, CUDA
Run, compile, and link C programs instantly with CJIT.
How wanting more is not always better
Microsoft is losing Builders fast. They're switching to MacOS and Linux. The biggest pull keeping people on Windows, outside of shear inertia, is content creation and gaming. However, even these are falling to Linux.
Rust struct serialization problem caused by many nullable SQL columns.
A question like this recently came up again. I once tried to find an answer, but only posted it in some "forum" a decade ago, and I think it's worth saving this little bit of trivia from digital nirvana.
This summary of Brad’s HIPS 2025 keynote looks at how HPC programming has (and has not) changed over the past 30 years
The FreeBSD Release Engineering Team is pleased to announce the availability of FreeBSD 14.5-RELEASE. This is the sixth release of the stable/14 branch.
Sometimes an article starts with a simple question or a single message on one of the available social platforms. This is the case today with this article where @ooberober asked: How small can the r…
A practical guide to speculative decoding in vLLM on AMD GPUs, covering draft-and-verify mechanics, MTP, EAGLE-3, DFlash, DSpark, configuration, tuning, and ben
fresh OI install, clean slate, let's see how well the updated OI Sun Ray support works
Eight long held and common beliefs about the network have been shown, time after time, to be false. What are they, and what do they mean?
A worked out example of different job queueing semantics and how they affect workloads.
When I was a kid , computers weren’t coddled and left running 24/7, when you were done with them you switched them off, and when you wanted them again you just switched them on and within a second or so they’d be loading whatever was in their disk drive.
