HackerTube

Rust at speed — building a fast concurrent database

Jon Gjengset2 HN recommenders · 2019–2022
225K views · lecture · evergreen · Uploaded 2019-01-05

A guest lecture Jon Gjengset gave at Two Sigma on using Rust for large, high-performance systems. It covers why Rust suits such projects, the Noria research database, a concurrent data structure used in it, and practical experience.

HN points to it as a detailed, captivating talk by the Noria author, including a good explanation of deduplicating data between maps using unsafe.

C, C++, Rust, Go & systems languagesDatabases & storageSoftware performance & optimization

Watch on YouTube

4 comments from Hacker News

@dpezely2019 · in “Rust at speed – building a fast concurrent database [video]”

He explains this at roughly the 40 minute mark. [1] "There's 7 more lines of `unsafe` that avoids keeping both maps. You still keep two maps, but you de-duplicate the data between those two maps." [VIDEO]

CitesView on HN
@harikb2020 · in “There’s plenty of room at the Top: computer performance after Moore’s law”

Anecdotal - One of favorite YouTube live-streaming personalities, Jon Gjengset[1] is a Ph.D student at MIT CSAIL. His talks are very detailed[4] and captivating[3] [1] [LINK] Talks [2] [VIDEO] [3] [LINK] Rust Details(my favorites) [4] [LINK]

RecommendsView on HN
@Orou2021 · in “Noria: Dynamic, partially-stateful data-flow for high-perf web applications”

The author of Noria, Jon Gjengset, gave a great talk about the system: [VIDEO] The speeds achieved are impressive but you do not get consistency.

RecommendsView on HN
@Jonhoo2022 · in “How PlanetScale Boost serves SQL queries faster”

:wave: Author of the paper this work is based on here. I'm so excited to see dynamic, partially-stateful data-flow for incremental materialized view maintenance becoming more wide-spread! I continue to think it's a _great_ idea, and the speed-ups (and complexity reduction) it can yield are pretty immense, so seeing more folks building on the idea makes me very happy. The PlanetScale blog post references my original "Noria" OSDI paper ( [LINK] ), but I'd actually recommend my PhD thesis instead ( [LINK] ), as it goes much deeper about some of the technical challenges and solutions involved. It also has a chapter (Appendix A) that covers how it all works by analogy, which the less-technical among the audience may appreciate :) A recording of my thesis defense on this, which may be more digestible than the thesis itself, is also online at [LINK] , as well as a shorter talk from a few years earlier at [VIDEO] . And the Noria research prototype (written in Rust) is on GitHub: [LINK] . As others have already mentioned in the comments, I co-founded ReadySet ( [LINK] ) shortly after graduating specifically to build off of Noria, and they're doing amazing work to provide these kinds of speed-ups for general-purpose relational databases. If you're using one of those, it's worth giving ReadySet a look to get these kinds of speedups there! It's also source-available @ [LINK] if you're curious.

CitesView on HN