Chapter 32. Taking the Plunge into GPU Computing

Ian BuckStanford University

This and many other chapters in this book demonstrate how we can use GPUs as general computing engines, but it is important to remember that GPUs are primarily designed for interactive rendering, not computation. As a result, you can run into some stumbling blocks when trying to use the GPU as a computing engine. There are significant differences between the GPU and the CPU regarding both capabilities and performance, ranging from memory bandwidth and floating-point number representation to indirect memory access. Because our goal in porting applications to the GPU is to outperform the CPU implementations, it is important to understand these differences and structure our ...

Get GPU Gems 2: Programming Techniques for High-Performance Graphics and General-Purpose Computation now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.