Perf-book is an extensive open-access book by Denis Bakhvalov and collaborators, covering how to analyze, understand, and improve performance on modern CPU architectures. It addresses topics such as CPU microarchitecture, memory hierarchy, data locality, branch prediction, vectorization, etc., with both theory and practical examples. It is written in LaTeX (with TeX/LaTeX source), augmented by tooling to export PDF, build with Pandoc, and with supporting materials like images, code, and exercises. The book is licensed under CC0 (public domain).
Features
- Covers CPU microarchitecture, caches, memory hierarchies, branch predictors, vector units, and related performance topics
- Provides practical tuning advice (code examples, benchmarking, profiling)
- Source files in LaTeX plus tooling: Python script for export, tools like pandoc, bibtex, pdflatex
- Glossary, metadata, figures, charts to aid understanding; visual aids (images, diagrams)
- Supports building PDF output on Windows and Linux (macOS possible but with extra steps)
- Freely distributable under a permissive (CC0) license; readers have full access to all content
Categories
SystemLicense
Creative Commons Attribution LicenseFollow Perf Book
nel_h2
AI-powered service management for IT and enterprise teams
Give your IT, operations, and business teams the ability to deliver exceptional services—without the complexity. Maximize operational efficiency with refreshingly simple, AI-powered Freshservice.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of Perf Book!