Topics /
The most popular blog posts about systems & performance on Hacker News. 2,560 posts.
Many products at Cloudflare aren’t possible without pushing the limits of network hardware and software to deliver improved performance, increased efficiency, or novel capabilities such as soft-unicast, our method for sharing IP subnets across data centers. Happily, most people do not need to know the intricacies of…
Recently I’ve been deliberately working on some areas of Firefox I’m unfamiliar with, particular relating to graphics. This led me to rewriting Firefox’s BMP decoder and learn a number of interesting things along the way.
On May 21st 2019, we broke search service out of vehicle-service. It was faster, cheaper, and better!
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.
Transformers have grown deeper and wider, but training them on long sequences remains difficult. The attention layer at their heart is the compute and memory bottleneck: doubling the sequence length would quadruple the runtime and memory requirements.
Microsoft has announced that it is ending the ability to cross-sign drivers, effective 1 July 2021. This will effectively make it impossible to release new or updated drivers for Windows 7, Windows…
This vulnerability allows remote attackers to execute arbitrary code on affected installations of Linux Kernel. Authentication is not required to exploit this vulnerability, but only systems with ksmbd enabled are vulnerable.
Textual is a TUI framework for Python, inspired by modern web development.
With git, a cryptographic signature on a commit provides strong integrity guarantees of the entire history of that branch going backwards...
UC Santa Cruz researchers show that it is possible to eliminate the most computationally expensive element of running large language models, called matrix multiplication, while maintaining performance.
RAD Game Tools' web page. RAD makes Bink Video, the Telemetry Performance Visualization System, and Oodle Data Compression - all popular video game middleware.
Discussions of coroutines and user-mode threads — like Project Loom's virtual threads or Go's goroutines — frequently turn to the subject of performance. The question I'll try answering here is, how do user-mode threads offer better application performance than OS threads?
Read how we built a flexible, high-performance columnar compression mechanism for PostgreSQL to add scalability to our favorite database.
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.
One of the headline features of CockroachDB is its full support for ACID transactions across a distributed database. In this post, we’ll explain...
At ISC 2024, Intel announces Aurora is the fastest AI supercomputer, has broken the exascale barrier, and details the importance of an open ecosystem in HPC and AI.
Rust struct serialization problem caused by many nullable SQL columns.
This summary of Brad’s HIPS 2025 keynote looks at how HPC programming has (and has not) changed over the past 30 years
News and information from Gentoo Linux
We estimate the magnitude of speed improvement for optimization solvers and computer hardware in the 35 years from 1989 to 2024.
