OpenCL Programming Guide

The definitive reference and programming guide for OpenCL, the new parallel programming standard for high performance and cloud environments. Will catch a growing wave of interest in OpenCL, following its implementations by Apple, NVIDIA, and ATI. Designed specifically for developers: shows how to write parallel OpenCL software that takes advantage of all available processors (CPUs, GPUs, accelerators, and more) OpenCL insiders present its API, programming language, development best practices, and many practical examples. The new OpenCL (Open Computing Language) standard and framework will make it possible to write applications that access all the programming resources available to them, including CPUs, GPUs, accelerators such as IBMs CELL, and even external processors. Already implemented by Apple (in Snow Leopard), ATI, and NVIDIA, OpenCL will be a solution for personal computers, servers, handheld/embedded devices, High Performance Computing applications, and even cloud systems. However, until now, developers who want to learn OpenCL have been required to read a specification designed for hardware and software implementers, not programmers. OpenCL Programming Guide is the first comprehensive, authoritative guide to OpenCL specifically designed for working developers. Written by leaders in the OpenCL community, this book thoroughly explains how each element of OpenCL 1.1 works, and offers complete reference material on the OpenCL API and programming language. The authors review key use cases, demonstrate how to write performant OpenCL code, and introduce essential algorithms for developers. They also present many examples, performance tips, case studies, and problems designed to help developers rapidly become effective, whatever their application or computing environment.