Book Image

C++ High Performance - Second Edition

By : Björn Andrist, Viktor Sehr
5 (2)
Book Image

C++ High Performance - Second Edition

5 (2)
By: Björn Andrist, Viktor Sehr

Overview of this book

C++ High Performance, Second Edition guides you through optimizing the performance of your C++ apps. This allows them to run faster and consume fewer resources on the device they're running on without compromising the readability of your codebase. The book begins by introducing the C++ language and some of its modern concepts in brief. Once you are familiar with the fundamentals, you will be ready to measure, identify, and eradicate bottlenecks in your C++ codebase. By following this process, you will gradually improve your style of writing code. The book then explores data structure optimization, memory management, and how it can be used efficiently concerning CPU caches. After laying the foundation, the book trains you to leverage algorithms, ranges, and containers from the standard library to achieve faster execution, write readable code, and use customized iterators. It provides hands-on examples of C++ metaprogramming, coroutines, reflection to reduce boilerplate code, proxy objects to perform optimizations under the hood, concurrent programming, and lock-free data structures. The book concludes with an overview of parallel algorithms. By the end of this book, you will have the ability to use every tool as needed to boost the efficiency of your C++ projects.
Table of Contents (17 chapters)
15
Other Books You May Enjoy
16
Index

Performance guidelines

I cannot stress enough the importance of having a concurrent program running correctly before trying to improve the performance. Also, before applying any of these guidelines related to performance, you first need to set up a reliable way of measuring what you are trying to improve.

Avoid contention

Whenever multiple threads are using shared data, there will be contention. Contention hurts performance and sometimes the overhead caused by contention can make a parallel algorithm work slower than a single-threaded alternative.

Using a lock that causes a wait and a context switch is an obvious performance penalty, but what is not equally obvious is that both locks and atomics disable optimizations in the code generated by the compiler, and they do so at runtime when the CPU executes the code. This is necessary in order to guarantee sequential consistency. But remember, the solution to such problems is never to ignore synchronization and therefore introduce...