The latest blog posts to take off on Hacker News. 6,942 posts.
SIMD has a reputation for being complex. I've met many very good software engineers who dismiss it as something too complex to learn or a niche optimization meant for only the highest-performance software, not useful in everyday programming.
I get kidnapped by the MIG while adding DDR3 to our CPU
Designing the world's first open source Ethernet switch ASIC. An unmanaged cut-through 3-port 100Mbps switch fabbed on Global Foundries 180 nm.
The Wasm GC and exceptions proposals are both enabled by default in today’sWasmtime 47 release! We are excited to help bring more languages to WebAssemblyand everywhere that Wasmtime runs. Getting to this point involved large Wasmtimecha...
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…
Philosophy - Sciences - Geekery - Art - Life - Coaching - Fun < Simplify Everything
How to make 768 distinct Postgres servers look like 1 to your applications.
A storage appliance that shipped before AVX2 existed now runs Google's Gemma 4 26B in my basement, no GPU. Here's how we got a modern MoE model onto a CPU older than the instructions its inference engine assumed were there.
I built a device to measure end-to-end input latency, then used it to find out what actually moves the needle.
As I continue reading An Introduction to Parallel Programming, I cannot help but notice a connection between communication among processors, communication among human beings, and communication within the individual self....
KO | EN | Discussions on HN I still have a netbook I bought back in 2009. The ASUS Eee PC 1000HE has an Intel Atom N280 and 1GB of DDR2 RAM. The Atom N series was Intel’s line of cheap, low-performance CPUs for netbooks. The chip has 56KB of L1 cache and 512KB of L2 cache, and its clock speed is 1.6...
A worked out example of different job queueing semantics and how they affect workloads.
How wanting more is not always better
Decommissioned enterprise tesla GPUs are widely available and extremely cheap. In this post, I benchmark tesla GPUs to see if they are useful
So I was optimizing a domain-specific compressor the other day, as one does. One important problem was chunking the input string and optimally choosing the most compact encoding for each chunk (different encodings compress different characters better, so where to split is not immediately obvious). The previous post…
Recently, while working on an optimized Quicksort implementation, I came across a rather interesting quirk. Modern compilers (especially Clang) optimize loops using fast, branch-free instructions - provided you use the right programming style.
If it quacks like a duck...
GhostLock (CVE-2026-43499) is a Linux kernel vulnerability found by Nebu that exists in every major distribution since 2011. Triggering the bug does not require any special kernel config or privilege. By turning it into a 97% stable privilege escalation and container escape, Google has rewarded us $92,337 in…
Cloudflare Research is building a global consensus service called Meerkat that uses a new consensus algorithm called QuePaxa. We plan to use Meerkat to build a strongly consistent, fault-tolerant key-value store, and other applications.
If you need a basic NAS and don't care about GUI features, it is suprisingly simple to set up a ZFS dataset and share it over the network using Samba.
PgDog is a connection pooler, load balancer, and sharding proxy for PostgreSQL. Scale Postgres horizontally without rewriting your application.
An end-to-end engineering practice for the MiMo-V2.5 series inference system, covering KVCache management, tiered caching, SWA-aware prefix cache trees, scheduling, Prefill/Decode pipelines, and multimodal optimizations.
Disclaimer: An earlier version of this post claimed the structure is wait-free, this is incorrect. Being wait-free requires that failure or suspension of any thread can’t cause failure or suspension of another thread. This queue in fact does not fulfill that requirement. The main section which discusses the wait…
The challenges around porting Linux to an old videogame console by Atari
Goodbye, old friend. We’ve been traveling together since 2010, and you’ve faithfully served all my virtual machine needs well, first with VirtualBox and then later with libvirt and KVM . But, as they 1 say, all good things must pass.
Published: 05-07-2026 20:33 | Author: Remy van Elst | Text only version of this article
I spent a whole month researching FreeBSD's kernel's virtual memory system to understand why tools like htop, btop and fastfetch report different information about RAM usage. I ended up submitting patches to all of these projects.
In this blog post, we will explain how strict memory overcommit protects your database from catastrophic OOM (out of memory) kills. We will also share how a one character kernel bug forced us to temporarily disable this setting. Finally, we will explain our heuristic for determining the right memory overcommit limit…
Keeping workflow state and application data in the same Postgres database lets one transaction update both, simplifying idempotency and atomicity.
