HackerTube

THE VIDEO SHELF · hash table design

CppCon 2017: Matt Kulukundis “Designing a Fast, Efficient, Cache-friendly Hash Table, Step by Step”

Matt Kulukundis explains how Google designed SwissTable, a fast, cache-friendly hash table, step by step from std::unordered_map. The talk covers open addressing, SIMD probing and the tradeoffs of changing a fundamental data structure.

Why HN readers recommended it

HN repeatedly cites it as the definitive explanation of SwissTable, the design behind Abseil, Rust's hash map and Go's maps.

This is a summary from the source dataset, not an endorsement by Hacker News or the video creator.

Video & recommendation details

Creator / channel
CppCon
Topic
hash table design
Format
talk
Level
advanced
HN recommenders in dataset
9
HN mentions in dataset
19
Recommendation years in dataset
2018–2026

These counts and years describe the included Hacker News records; they are not YouTube views or the video’s publication date.

From the original HN conversations

Excerpts from the recorded comments. Follow each source for the full comment and surrounding discussion.

  • I think you're underestimating the number of memory redirections that are necessary for chaining. It's more than one. That's the killer. It's important to be wary of assuming your data will be in cache. It usually isn't. The problem isn't making lots of high density …

    nwallin on Hacker News · 2019-10
    In Hash tables
  • Very cool! Independent of the cool use of `aesenc` and `aesdec`, the features for skipping ahead in the random stream and forking a separate stream are awesome. > My current home project is bump-allocator + semi-space garbage collection in SIMD for GPUs. As far as …

    sujayakar on Hacker News · 2021-03
    In MIPS Becomes RISC-V
  • Related: this talk [VIDEO] about the making of absl::flat_hash_map (a "competitor" to boost::unordered_flat_map) is one of my favorites.

    KerrickStaley on Hacker News · 2022-11
    In Inside boost::unordered_flat_map

Summaries, classifications and recommendation records are from the HackerTube source dataset. Read about this library and its limitations.