Overview
Dive deep into C++ performance optimization in this CppCon 2022 conference talk. Explore algorithmic complexity, data locality, parallelism, and compiler optimizations, with a focus on concurrency. Learn to select efficient algorithms and data structures, analyze their impact on program performance, and understand the importance of data locality. Examine tradeoffs between space and time complexity, as well as setup-time versus run-time considerations. Gain insights into hardware caches, cache models, cache lines, and cache misses through practical examples. Discover techniques for optimizing object-oriented code and implementing cache-oblivious algorithms. Acquire guidelines for writing performant C++ code and learn effective benchmarking practices to ensure your programs stay ahead of the game in terms of efficiency.
Syllabus
Intro
Hardware
Caches
My Machine
Cores
Latency
Cache Models
Cache Lines
Cache Misses
Cache Miss Example
False Sharing
Cache Line Example
Main Cache
Memory
Summary
ObjectOriented Example
Cash oblivious
Guidelines for code
Benchmarking
Taught by
CppCon