Talks to Watch

"Compacting the Uncompactable" by Bobby Powers

29K views · Sep 14, 2019 · 40:18 · 3 HN recommenders · 4 mentions · discussed 2020–2025

Bobby Powers presents Mesh, a drop-in malloc replacement that eliminates memory fragmentation in unmodified C/C++ programs by compacting memory. The talk covers how it works despite C's fixed object addresses, and its benefits for programs like Redis.

Why Hacker News points people here: Recommended as a great talk on solving fragmentation for C/C++ and runtimes like Ruby and Python without code changes.

“There was a great talk at Strange Loop about a drop-in malloc replacement which compacts.”HN comment, 2022

What Hacker News said 4

RecommendsremexreDec 2020

“how to make a malloc() and free() that perform compaction”

Talk on a malloc that compacts memory to return it to the OS.

Full comment

There's an interesting talk [0] from Bobby Powers on how to make a malloc() and free() that perform compaction, to be able to release memory back to the operating system more frequently (and improve cache hit rates, etc.). But this isn't standard at all yet. [1]: this video

in “How the Python object system works” · view on HN
RecommendsrtfeldmanJul 2022

“There was a great talk at Strange Loop about a drop-in malloc replacement which compacts.”

Drop-in compacting malloc replacement that improved memory use in projects like Redis.

Full comment

There was a great talk at Strange Loop about a drop-in malloc replacement which compacts. Apparently it actually led to memory usage improvements in industrial projects like Redis: this video

in “Reference count, don't garbage collect” · view on HN
RecommendssuperjanJan 2025

“also a good presentation about it on youtube.”

Presentation on merging sparse memory pages, a kind of garbage collection for C/C++.

Full comment

There’s a great paper from the group of Emery Berger, where they merge memory pages that become sparse. A kind of garbage collection for C and C++ languages. There’s also a good presentation about it on youtube. link The presentation: this video

in “Using black magic to make a fast circular buffer (2017)” · view on HN
Cites itssokolowJun 2022

“there's been some interesting progress on that front”

Shows progress on reducing fragmentation overhead in malloc/free systems.

Full comment

> but malloc/free based systems also have a relatively high memory overhead due to fragmentation In case you haven't seen it, there's been some interesting progress on that front: this video

in “Rust Is Hard, Or: The Misery of Mainstream Programming” · view on HN