Operating systems and hardware 0/198

Operating systems and hardware

Caches and performance

The memory hierarchy, cache lines, locality and false sharing, and how to measure latency without fooling yourself. Most of the speed in a data structure comes from how it touches memory.

0 / 29 solved

Reading

WhiteBox lessons1

Papers and essays1

  1. Latency Numbers Every Programmer Should Know(opens in a new tab)

    Dean and Norvig 2012 Essay 10 min

    A table of typical costs, from an L1 cache reference (0.5 ns) and a main memory reference (100 ns) to a round trip within a datacenter (0.5 ms), a disk seek (10 ms) and a packet from California to the Netherlands and back (150 ms), credited to Jeff Dean and originally Peter Norvig and collected in Jonas Bonér's gist. The gist now also lists rough 2026 timings for local and hosted LLMs.

Videos3

Classic Talks on Performance and Production Various speakers 2 videos

Jane Street Tech Talks: Systems Jane Street 1 video

People to follow2

  • Casey Muratori

    Computer, Enhance!

    Courses on how the CPU actually runs your code and why programs end up slow, starting from only some programming experience.

  • Brendan Gregg

    Creator of flame graphs

    Methods, tools and worked examples for finding where a slow Linux system spends its time.