151
Erlang ARM32 JIT is born
A modular embedded ecosystem—bare-metal hardware, software stacks, and a cloud platform for real-time IoT and distributed systems.
New
The latest blog posts to take off on Hacker News. 543 posts.
152
Under the hood: Vec<T>
Let's look at Vec<T> to get a better understanding of its inner structure.
153
CPU cache-friendly data structures in Go
False sharing, AoS vs SoA, random access, branch prediction: re-measured with go test -bench on two machines. Some claims held, some didn't.
154
Why we need SIMD
tl/dr: it can deliver big performance speedups at modest area cost
155
Memory access is O(N^[1/3])
In computer science, we often compare the efficiency of algorithms by describing their runtime as a function of the size of the input. Sorting is O(n * log(n)), meaning that sorting a list of N items takes an amount of time proportional to the number of items multiplied by its logarithm. Matrix multiplication is…
156
The QNX Operating System
Quantum Software and the microkernel UNIX
157
Matrix Core Programming on AMD GPUs
In this blog post, we walk through how to use Matrix Cores in HIP kernels, with a focus on low-precision data types such as FP16, FP8, and FP4, as well as the new family of Matrix Core instructions with exponent block scaling introduced in the AMD CDNA™4 architecture. Through code examples and illustrations, we…
158
Alibaba cloud FPGA: the $200 Kintex UltraScale+
Using a decommissioned Alibaba cloud accelerator card as an FPGA dev board
159
Zig builds are getting faster
Andrew Kelley famously (or infamously, depending on your views) said "the compiler is too damn slow, that's why we have bugs." 1
160
The G in GPU is for Graphics damnit
As is true for everything, a lot of things need to happen for anything to happen, and so it’s true for this blogpost as well. Out of all of these everything that needed to happen, 3 are these:
161
The best worst hack that saved our bacon
No-one really likes engineering war stories, but this one's relevant because there's a moral to it. I've talked before about defining technical debt as technical decisions that provide immediate value, but with long-term negative impact if they aren't cleaned up. Sometimes introducing technical debt
162
Intelligent Kubernetes Load Balancing at Databricks
How Databricks uses client-side load balancing to improve gRPC traffic, reduce tail latency, and handle ingress with real-time service discovery.
163
Mind the encryptionroot: How to save your data when ZFS loses its mind
While ZFS has a well-earned reputation for data integrity and reliability, OpenZFS native encryption has some incredibly sharp edges that will cut you if you don't know where to be careful. Unfortunately, I learned this the hard way, standing in a pool of my own blood and tears after thoroughly lacerating myself…
164
Safe zero-copy operations in C#
My attempt at talking about one of the most underrated features of C#.
165
Not all OCuLink eGPU docks are created equal
I recently tried using the Minisforum DEG1 GPU Dock with a Raspberry Pi 500+, using an M.2 to OCuLink adapter, and this chenyang SFF-8611 Cable. After figuring out there's a power button on the DEG1 (which needs to be turned on), and after fiddling around with the switches on the PCB (hidden under the large metal…
166
Optimizing a 6502 image decoder, from 70 minutes to 1 minute
A little walkthrough on an image decoder algorithm's high-level optimization, in order to make more digestable for a 6502 processor.
167
We bought the whole GPU, so we're damn well going to use the whole GPU
Benjamin Spector* , Jordan Juravsky* , Stuart Sul* , Dylan Lim , Owen Dugan , Simran Arora , Chris Ré
168
VMScape and why Xen dodged it
ETH Zürich’s new VMScape attack hit KVM and VMware, but Xen’s microkernel-like design kept it out of reach.
169
The Weird Concept of Branchless Programming
Read THE WEIRD CONCEPT OF BRANCHLESS PROGRAMMING on sanixdk.xyz
170
Beyond OpenMP in C++ and Rust: Taskflow, Rayon, Fork Union
Most C++ and Rust thread pools run 10x slower than OpenMP on fork-join workloads. Fork Union closes the gap to 20% with 300 lines, dodging mutexes and CAS.
171
How does lossless compression in Fuji RAF files work? (2020)
I spent 3 weeks earlier this year digging in to and reimplementing the lossless compression algorithm in Fuji’s RAF Camera RAW files. This post discusses the fundamentals of how it works.
172
Fast UDP I/O for Firefox in Rust
Motivation Around 20% of Firefox’s HTTP traffic today uses HTTP/3, which runs over QUIC, which in turn runs over UDP. This translates to substantial UDP I/O activity. Firefox uses NSPR for most of its network I/O. When it comes to UDP I/O, NSPR only offers a limited set of dated APIs, most relevant here PR_SendTo and…
173
Some interesting stuff I found on IX LANs
These days the internet as a whole is mostly constructed out of point to point ethernet circuits, meaning an ethernet interface (mostly optical) attached directly from one routing device to another routing device.
174
How fast is Go? Simulating particles on a smart TV
The challenge, simulate millions of particles in golang, multi-player enabled, cpu only, smart tv compatible.
175
Why is Windows still tinkering with critical sections? – The Old New Thing
The critical section may be an old dog, but it's still learning new tricks.
176
Preparing for the .NET 10 GC
In .NET 9 we enabled DATAS by default. But .NET 9 is not an LTS release so for many people they will be getting DATAS for the first time…
177
How AWS S3 serves 1 petabyte per second on top of slow HDDs
Learn how Amazon built the backbone of the modern web that scales to 1 PB/s and 150M QPS on commodity hard drives
178
179
180
Wild performance tricks
Last week I had the pleasure of attending RustForge in Wellington, New Zealand. I gave a talk titled “Wild performance tricks”. You can watch a recording of my talk . If you’d prefer to read rather than watch, the rest of this post will cover more or less the same material. The talk shows some linker benchmarks, which…