mshah.ioCourse ScheduleCourse Information
Administrivia, Course Introduction, Systems Overview (Shah & Zhong)C Review and Unix Programming Tools (Shah)No ClassTransistors-instructions (Zhong)Architecture (Zhong)Assembly 1: Machine Instructions, and Reverse Engineering (Shah)Assembly 2: Assembly and Machine Representation(Shah)Address Space and Abstract Machine (Zhong)Abstract machine: implementation and Operating System: Process (Zhong)Operating System: File (Zhong)OS: Virtual Memory (Zhong)Memory Hierarchy (Zhong)Locality and the Cache (Zhong)(Lin) continued on Cache and Compilers 1 (Shah)Linkers and Code Generation (Shah)No ClassCompilers 2 - Optimization (Shah)ExamOS: Heap Allocators 1 (Shah)OS: Heap Allocators 2 (Shah)Modern Systems Programming: D and Memory Safety (Shah)Modern Systems Programming: Rust and Memory Safety (Zhong)mini - Operating Systems, xv6 (Shah)xv6 and Schedulers and operating system multiprocessing (Shah)CPU Parallelism with SIMD (SSE, AVX), Concurrency and Threads (Shah)GPU Programming (Zhong)No ClassNo ClassGPU Programming and LLMs (Zhong)Course wrap-up, class photo, Instructors Choice of topic(e.g. Networking) (Shah)


Module 13 - Locality and the Cache (Zhong) - 2025-Oct-06

Overview

In this module we learn about the importance of undersetanding hardware for performance.

  • Locality
  • Caches

Slides and Class Content

Please note: Your in-class activity is graded. Your first attempt will be graded. You can use your notes and slides if you like. You have 1 week from the date of this module opening to complete the in-class activity. In-class activities are completed individually.

Audio/Video Recording

Note: Audio/Video recordings are made in a best effort in case you have to miss class or you would otherwise like to re-watch a portion of lecture. Occassional recording errors (e.g. missing sound, corrupted video, etc.) may occur, and I apologize in advance if a recording is not able to be made available.

Readings

    Please use the following resources to supplement this lecture. You may choose to read 1 or more of these before/during/after the lecture to fully immerse into the material.

  1. Local (arena) Memory Allocators - John Lakos [ACCU 2017]
    • Video on memory allocators with some segments explaining basics of caching
  2. code::dive conference 2014 - Scott Meyers: Cpu Caches and Why You Care]
    • Video on memory allocators with some segments explaining basics of caching
  3. CppCon 2016: Timur Doumler 'Want fast C++? Know your hardware!
    • Video on caching
  4. Caches, Video 6: Code optimization for caches
    • Video on Cache miss analysis
  5. Cache-Oblivious Algorithms and Data Structures
    • Cache Oblvious Algorithms by Erik Demaine
  6. The Working Set Model for Program by Peter J. Denning
    • Paper describing that the amount of resources we need as related to cache in the 'working set'
Please do not redistribute or host any materials without e-mailing me first. Thank you for your time!