Heterogeneous Computing With Opencl

by Gaster; Howes; Kaeli; Mistry; Schaa

• ISBN13: 9780123877666
• ISBN10: 0123877660
• Format: Paperback
• Copyright: 2011-08-17
• Publisher: Elsevier Science

Summary

Heterogeneous systems that incorporate CPUs and GPUs offer increased performance, flexibility and efficiency, and facilitate faster results and more detailed simulations than were possible only a few years ago. Derived from familiar C/C++ code, OpenCL lets you leverage these systems by assigning each processor the task it is best suited for, accessing massively parallel GPUs for computationally-intensive tasks. Written by leaders in the parallel computing and OpenCL communities, Heterogeneous Computing with OpenCL will give you hands-on OpenCL experience to address a range of fundamental parallel algorithms.  Detailed examples illustrate the power and elegance of OpenCL to handle image processing, web plugins, random number generation, video editing, and more. The authors explore multiple device architectures: CPUs, GPUs, and fully-integrated Accelerated Processing Units (APUs) such as AMD Fusion. Demonstrates OpenCL and parallel programming on CPUs, GPUs, and Accelerated Processing Units (APUs) Provides a grounding in data, task, and thread-level parallel computing Covers memory spaces, optimization, timing, and debugging Includes multiple examples and case studies you can apply to your own projects