Lock-Free Rust: How to Build a Rollercoaster While It’s on Fire
Lock-Free Rust: How to Build a Rollercoaster While It’s on Fire. “Every stitch in that flag was a commitment to thread safety without locks. std::atomic was the needle.” — Betsy Ross, Federalist Papers, Draft 29 (suppressed), 1782 Buckle your seatbelts, grab a helmet, and say goodbye to your loved ones because over the next few minutes, I am going to teach you more about lock-free data structures
Succinct data structures
Succinct data structures By Martijn Faassen • 2025-02-28 • Tags: computer science, xml, rust Introduction A few months ago, searching for ideas on how to make some code faster, I found myself reading a bunch of computer science papers. I don't pretend to be good at this - but I don't mind some confusion or being overwhelmed, and I'm okay to admit my ignorance. 1. I ran into this 15 year old paper
Solving the ABA Problem in Rust with Tagged Pointers
Introduction 📺 Series Overview 👨🔬 What Is the ABA Problem? 🏷️ Tagged Pointers with Versioning ⚙️ Key Mechanism 🧐 Pros and Cons 🧪 Tests and Benchmarks Tests Module Benchmarks Module 🔗 Resources 🤔 Final Thoughts Introduction Concurrent programming is challenging, especially when designing lock-free data structures. One subtle issue that can arise is the ABA problem, leading to unexpected be
Rope Optimizations, Part 1 - Zed Blog
A couple of weeks ago I came across one of Antonio's PRs, titled "Speed up point translation in the Rope" — now who doesn't stop to take a closer look at a PR with that title? The description already lives up to the title. It contains benchmark results telling me that a method on our Rope named point_to_offset is now up to 70% faster. 70%! Throughput increased by 250%. (I can't remember but I'm su
Designing A Fast Concurrent Hash Table
I recently released papaya, a fast and feature-complete concurrent hash table for Rust. In this post I want to dive into the design and research that went into creating it, as well as why you might consider using it over existing solutions. If you're looking for an overview of papaya, you might find it more useful to consult the documentation. Philosophy Concurrent hash tables are a well explored
Taming Floating-Point Sums | orlp.net
Suppose you have an array of floating-point numbers, and wish to sum them. You might naively think you can simply add them, e.g. in Rust: fn naive_sum(arr: &[f32]) -> f32 { let mut out = 0.0; for x in arr { out += *x; } out } This however can easily result in an arbitrarily large accumulated error. Let’s try it out: naive_sum(&vec![1.0; 1_000_000]) = 1000000.0 naive_sum(&vec![1.0; 10_000_000]) = 1
Rope & SumTree - Zed Blog
For this second post in Zed Decoded, our blog & video series in which we're taking a closer look at how Zed is built, I've talked to Zed's three co-founders — Nathan, Max, Antonio — about the data structure at the heart of Zed: the rope. This post comes with a 1hr companion video, in which Thorsten, Nathan, Antonio, and Max use Zed to look at how Zed uses the Rope and SumTree types. It's a loose c
Trying to invent a better substring search algorithm - moao's blog
To start, calculate the sum of the first window naively: First sum is 15. For the rest, reuse the previous sum, subtract the removed element and add the new one: Second = First + 6 - 0 = 21. Third = Second + 7 - 1 = 27. Fourth = Third + 8 - 2 = 33. Fifth = Fourth + 9 - 3 = 39. And so on... That's O(N) for all windows, and addition is, of course, fast as hell. The algorithm Let's preprocess the sum
PentominoをBitboardを使って解く - すぎゃーんメモ
子どもが百均で買ってきたパズルをやってみたら、全然うまく出来なくて悔しかったのでプログラムで解を探すことにした。 Pentominoとは 探索アルゴリズムと実装 計算量概算 効率的な探索の方針 backtracking Bitboardによる検索と判定 候補の事前計算 実装と実行結果 反転・回転での重複の除外 高速化 短辺から埋める X を最初に配置する Bitboardの反転/回転 Delta Swap による手法 巨大テーブルによるループ効率化 もう一方のループ効率化 孤立点検出による枝刈り その他の実装・手法 WebAssembly化してWebAppで使う 参考 まとめ Pentominoとは 調べるまで知らなかったが、このような 5つの正方形を繋げて作られる12種類のピースを使って敷き詰める、というパズルを Pentomino(ペントミノ) というらしい。 en.wikipedia
Fast(er) binary search in Rust
Intel Core i7-1068NG7 CPU @ 2.30GHz32Gb LPDDR4X 3733 MHzDarwin Kernel Version 22.4.0 In practical terms, the performance of binary search starts to drop when the dataset becomes larger than the size of the on-core cache (L2), which in my case is 512KB or 128K u32 elements. Eytzinger Layout Link to headingOne way to address the memory access predictability problem is to use the Eytzinger layout. It
Building a Fibonacci Heap — Kurt Lawrence
A Fibonacci heap is a heap data structure which leverages laziness to obtain a few favourable time complexities. A Fibonacci heap is used as a priority queue, usually used in graph algorithms to find the shortest path, such as A* or Dijkstra's algorithm. Compared to a binary heap, a Fibonacci heap has an amortised time complexity for inserts and key decreases of O(1) versus a binary heap's O(log n
The World's Smallest Hash Table | orlp.net
This December I once again did the Advent of Code, in Rust. If you are interested, my solutions are on Github. I wanted to highlight one particular solution to the day 2 problem as it is both optimized completely beyond the point of reason yet contains a useful technique. For simplicity we’re only going to do part 1 of the day 2 problem here, but the exact same techniques apply to part 2. We’re go
diziet | Stop writing Rust linked list libraries!
tl;dr: Don’t write a Rust linked list library: they are hard to do well, and usually useless. Use VecDeque, which is great. If you actually need more than VecDeque can do, use one of the handful of libraries that actually offer a significantly more useful API. If you are writing your own data structure, check if someone has done it already, and consider slotmap or generation_arena, (or maybe Rc/Ar
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
Integer Compression
K-Means Image Compression
A Lock-Free Vector
What is HyperLogLog and how to build yours in Rust
Build your own CountMinSketch in Rust
Build your own Counting Bloom Filter in Rust
GitHub - orlp/glidesort: A Rust implementation of Glidesort, my stable adaptive quicksort/mergesort hybrid sorting algorithm.
