Hacker News with Generative AI: Performance

What Do Van Halen and Brown M&M's Have to Do with Safety? (safetydimensions.com.au)
There’s a long tradition of musicians and actors adding absurd demands to their performance contracts just because they could.
Music, Performance, Contracts, Safety
9 points by mooreds 5 hours ago | 1 comments
Fast Video Generation with Sliding Tile Attention (hao-ai-lab.github.io)
TL;DR: Video generation with DiTs is painfully slow – HunyuanVideo takes 16 minutes to generate just a 5-second video on an H100 with FlashAttention3. Our sliding tile attention (STA) slashes this to 5 minutes with zero quality loss, no extra training required. Specifically, STA accelerates attention alone by 2.8–17x over FlashAttention-2 and 1.6–10x over FlashAttention-3.
Video Generation, Artificial Intelligence, Computer Vision, Optimization, Performance
12 points by zhisbug 2 days ago | 2 comments
RPython Compilation Quick Reference (tilde.town)
Well first things first, this is way more ram hungry than you'd expect, and you're restricted to python 2. BUT, you do end up with nice fast binaries at the end. So let's proceed!
Python, Programming, Performance, Compilers
3 points by amcclure 2 days ago | 0 comments
Kafka at the low end: how bad can it get? (broot.ca)
There is oft-quoted advice that Kafka does poorly as a job queue. I’ve experienced this myself, and I wanted to formalize it a bit.
Kafka, Performance, Queues, Software Engineering
125 points by alexwebr 2 days ago | 117 comments
One year after switching from Java to Go (glasskube.dev)
I always told people memory is cheap, black magic is OK and efficiency doesn't matter in most cases, but boy, how wrong was I...
Programming Languages, Go, Java, Performance, Software Development
252 points by pmig 2 days ago | 475 comments
Making my debug build run 100x faster so that it is finally usable (gaultier.github.io)
SIMD and dedicated silicon to the rescue.
Optimization, Performance, Debugging, Software Development
82 points by broken_broken_ 2 days ago | 20 comments
Nvidia GeForce GTX 980 Through GeForce RTX 5080/5090 GPU Compute Performance (phoronix.com)
Complementing the recent Linux GPU benchmarks of the NVIDIA GeForce RTX 5080 and GeForce RTX 5090 looking at both the Linux / Steam Play gaming performance as well as GPU compute and other areas, in today's testing is a wide multi-generation look seeing how the NVIDIA GeForce performance has evolved going back to the GeForce GTX 980 Maxwell GPUs up through the newest GeForce RTX 5080/5090 graphics cards.
Graphics Cards, Performance, Gaming, Linux, Benchmarks
27 points by mfiguiere 3 days ago | 15 comments
High-School Band Contests Turn Marching into a Sport–and an Art (newyorker.com)
Band kids today don’t just parade up and down the field playing fight songs. They flow across it in shifting tableaux, with elaborate themes and spandex-clad dancers.
Music, Art, Performance, Education, Culture
9 points by PaulHoule 3 days ago | 1 comments
The Impact of Metadata Configurations on Text-to-SQL Performance [pdf] (corraldata.com)
Text-to-SQL, Databases, Performance, Metadata
4 points by lkrubner 4 days ago | 0 comments
AVX-512 gotcha: avoid compressing words to memory with AMD Zen 4 processors (lemire.me)
The recent AMD processors (Zen 4) provide extensive support for the powerful AVX-512 instructions.
Processors, Optimization, AMD, Hardware, Performance
6 points by ibobev 6 days ago | 0 comments
Linux kernel cgroups writeback high CPU troubleshooting (dasl.cc)
We’ve been upgrading the operating system from CentOS to Ubuntu on hosts across our fleet. Our CentOS hosts run an outdated Linux kernel version (3.10), whereas our Ubuntu hosts run a more modern kernel version (6.8). In August 2024, we began rolling out the Ubuntu upgrade across our Apache web servers. When we migrated larger portions of our fleet to Ubuntu, we began seeing elevated listen overflow errors.
Linux Kernel, Troubleshooting, Performance, Operating Systems, Server Management
178 points by mesto1 6 days ago | 27 comments
Rust: Doubling Throughput with Continuous Profiling and Optimization (polarsignals.com)
“68.37% of CPU [...] with a one-line code change [...] went down to 31.82%”
Rust, Optimization, Performance, Profiling, Continuous Integration
59 points by mesto1 6 days ago | 34 comments
Tiny JITs for a Faster FFI (railsatscale.com)
Can we have a faster FFI for CRuby? Yes.
Ruby, Programming Languages, Performance, JIT Compilation
316 points by hahahacorn 8 days ago | 138 comments
The average CPU performance of PCs and notebooks fell for the first time (cpubenchmark.net)
Over 1,000,000 CPUs Benchmarked
Hardware, Performance, Trends, Computer Technology, Benchmarks
311 points by doener 8 days ago | 238 comments
Linux 6.13 Performance for 250Hz vs. 1000Hz Timer Frequency Comparison (phoronix.com)
Given the recent patch proposal to raise the Linux kernel's default timer frequency from 250Hz to 1000Hz, I ran some fresh benchmarks looking at the 250Hz vs. 1000Hz comparison on some modern desktop hardware.
Linux, Operating Systems, Performance, Benchmarks, Hardware
3 points by LorenDB 8 days ago | 1 comments
Show HN: Fluvio 38.8x faster than Kafka (infinyon.com)
At InfinyOn for the past 6 years we have obsessed over developer ergonomics, functionality, and reliability of Fluvio and Stateful DataFlow. It’s not trivial to build a distributed streaming engine from the ground up.
Software, Streaming, Performance, Comparisons, Open Source
10 points by debadyutirc 8 days ago | 2 comments
Do It Yourself Database CDN with Embedded Replicas (turso.tech)
Imagine you have a user in Singapore, and your database is in the US. Every time the user makes a request, it has to travel halfway around the world, which can lead to high latency and poor performance.
Database, Performance, Cloud, Networking, CDN
19 points by emschwartz 8 days ago | 2 comments
Git clone –depth 2 is vastly better than –depth 1 if you want to Git push later (stackoverflow.com)
I've done a shallow clone of a large repo (git clone --depth 1 [email protected]:myOrg/myRepo.git). I can push new changes to the remote but the first push is horribly slow. Subsequent pushes are fine. The command indicates that the first push writes a lot of data to the remote:
Git, Version Control, Optimization, Performance
137 points by jakub_g 8 days ago | 35 comments
The first yearly drop in average CPU performance in its 20 years of benchmarks (tomshardware.com)
Hardware, Benchmarks, Performance
149 points by LorenDB 9 days ago | 63 comments
Go's new map implementation in 1.24 is powered by Swiss Tables (twitter.com)
Go, Programming Languages, Data Structures, Performance
60 points by tosh 9 days ago | 4 comments
Can Cheap MiniPC with FreeBSD 14 Outperform MacBook Pro M1 Pro? (interfacecraft.online)
I put my €1800 MacBook Pro M1 Pro head-to-head with a €300 mini PC and found the cheaper option surprisingly fast.
Mini PCs, FreeBSD, MacBook Pro, Hardware, Performance
10 points by prmoustache 9 days ago | 6 comments
The Lengthiest HTTP Headers (fastly.com)
Large web page bodies make your page load slowly, but what about large headers?
Web Development, Performance, HTTP
4 points by HieronymusBosch 10 days ago | 0 comments
The missing tier for query compilers (scattered-thoughts.net)
Database query engines used to be able to assume that disk latency was so high that the overhead of interpreting the query plan didn't matter. Unfortunately these days a cheap nvme ssd can supply data much faster than a query interpreter can process it.
Databases, Query Compilers, Performance, Hardware, Cloud Computing
88 points by jamii 11 days ago | 9 comments
Scaling with PostgreSQL without boiling the ocean (shayon.dev)
“Postgres was great when we started but now that our service is being used heavily we are running into a lot of ‘weird’ issues”
Databases, PostgreSQL, Performance, Scaling
40 points by plaur782 11 days ago | 14 comments
JEP draft: 4-byte Object Headers (Experimental (openjdk.org)
Reduce the size of object headers in the HotSpot JVM from between 64 and 128 bits down to 32 bits on 64-bit architectures. This will reduce heap size, improve deployment density, and increase data locality.
Java, JVM, Performance, Optimization, Memory Management
8 points by zspitzer 11 days ago | 1 comments
No-Libc Zig Now Outperforms Glibc Zig (ziglang.org)
No-Libc Zig Now Outperforms Glibc Zig
Programming Languages, Performance, Zig
29 points by zdw 12 days ago | 1 comments
No-Libc Zig Now Outperforms Glibc Zig (ziglang.org)
No-Libc Zig Now Outperforms Glibc Zig
Programming, Zig, Performance
24 points by grayhatter 13 days ago | 10 comments
GCC 15 Compiler Showing Off Nice Performance Improvements on AMD Zen 5 (phoronix.com)
There were a number of other applications with small but consistent performance improvements when built by GCC 15.
GCC, Compilers, Performance, AMD, Zen 5
7 points by rbanffy 13 days ago | 1 comments
Solving Postgres' Search Limitations (paradedb.com)
We recently completed one of our biggest engineering bets to date: migrating pg_search, a Postgres extension for full text search and analytics, to Postgres' block storage system. In doing so, pg_search is the first-ever extension1 to port an external file format to Postgres block storage.
Databases, PostgreSQL, Search, Open Source, Performance
7 points by philippemnoel 13 days ago | 0 comments
From hours to 360ms: over-engineering a puzzle solution (danielh.cc)
In January 2025, Jane Street posted an interesting puzzle:
Programming, Optimization, Puzzles, Performance
97 points by max__dev 15 days ago | 35 comments