26 stories tagged performance-optimization, newest first.
268 points by stmonty | 94 comments
Richard Feldman's blog post benchmarking roc-deflate, a pure Roc DEFLATE library, against C's libdeflate, finding it under 10% slower, and describing Roc-specific optimizations.
9 points by fbuilesv | 2 comments
Rails at Scale blog post on speeding up Rails ActiveRecord enum methods by generating static code via module_eval instead of define_method and public_send, with benchmarks.
25 points by jstrieb | 0 comments
A 2014 blog post by Leandro Pereira on tia.mat.br about fast techniques for converting integers to strings.
66 points by ibobev | 7 comments
Playdate developer forum thread sharing advanced C optimization tricks learned while building a full-speed Game Boy emulator for the Playdate.
72 points by pptadversary | 11 comments
Blog post by Hayder Tirmazi describing optimizations that make prompt lookup drafting in llama.cpp up to 42x faster (140x with Daniel Lemire's follow-up PR) with less memory.
13 points by shishir03 | 3 comments
Medium post giving a simplified overview of a Paesani Research Group (UC San Diego) paper on automatically optimizing a molecular simulation program.
592 points by Jach | 442 comments
Dan Luu essay arguing that LLM coding agents make specialized performance optimization cheap, illustrated with optimizing a regex engine for specific workloads.
424 points by tosh | 91 comments
Blog post on using OpenAI Codex for automated research to optimize a kernel 232x faster; the captured page shows a Cloudflare verification screen.
568 points by WadeGrimridge | 211 comments
Mitchell Hashimoto blog post arguing SIMD is simple to learn and showing the common shape of SIMD code for speeding up loops.
344 points by vok | 152 comments
Daniel Lemire blog post arguing that binary search can be beaten, illustrated with quaternary search.
373 points by benjoffe | 87 comments
Ben Joffe's article presenting a 64-bit date conversion algorithm that counts years backwards and needs only 4 multiplications, with an open-source C++ implementation.
376 points by thunderbong | 148 comments
Blog post recounting how the author spent weeks hand-optimizing assembly for a Java platform, only to find the gains came from benchmarking on unrealistic random data.
341 points by harambae | 96 comments
Tom's Hardware article reporting FFmpeg developers achieving a roughly 100x speedup in a function via handwritten assembly code.
563 points by immortaljoe | 236 comments
GitLab engineering blog post explaining how fixing a 15-year-old Git function bottleneck cut repository backup times from 48 hours to 41 minutes.
387 points by mfiguiere | 182 comments
Stanford CRFM blog post by Anne Ouyang, Azalia Mirhoseini and Percy Liang describing surprisingly fast AI-generated GPU kernels they discovered.
443 points by jedeusus | 139 comments
Go Optimization Guide is a series of technical articles on patterns and techniques for writing high-performance Go applications.
638 points by wicket | 201 comments
Fabien Sanglard's blog post explaining the optimizations that make FastDOOM, a DOS port of Doom, run faster on 486-era PCs like the IBM PS/1.
585 points by atombender | 196 comments
A long blog post by Ragnar Groot Koerkamp implementing and optimizing a static S+ search tree for high-throughput sorted-data search, reaching up to 40x faster than binary search.
553 points by eaufavor | 156 comments
Cloudflare blog post on cutting Pingora proxy CPU usage by optimizing header handling in Rust with a custom trie data structure, trie-hard.
343 points by chris_overseas | 200 comments
arXiv paper presenting C++ design patterns and benchmarks for low-latency code in high-frequency trading, including a Disruptor pattern implementation.
368 points by weinzierl | 72 comments
GitHub pull request to llama.cpp by jart that upstreams llamafile's vectorized expf() functions to compute SiLU and softmax faster and more accurately on CPU.
403 points by nalgeon | 169 comments
Ben Hoyt's blog post walking through nine Go solutions to the One Billion Row Challenge, cutting runtime from 1m45s to 4s.
406 points by ingve | 70 comments
Blog post by Miguel Young de la Sota explaining how to design a fast, portable SIMD algorithm in Rust, using base64 decoding as the example.
400 points by udev4096 | 229 comments
A plain web page listing Rob Pike's five rules of programming on measurement, simple algorithms, and data structures, with notes on related maxims.
462 points by SEJeff | 87 comments
Blog post exploring how Linux pipes are implemented by iteratively optimizing a pipe-based test program from 3.5GiB/s to about 65GiB/s using perf profiling.
477 points by tentacleuno | 193 comments
A Code Golf Stack Exchange answer presenting an extremely optimized FizzBuzz implementation reaching about 55 GiB/s throughput.