CUDA by Example: An Introduction to General-Purpose GPU Programming

摘要: This book is required reading for anyone working with accelerator-based computing systems. From the Foreword by Jack Dongarra, University of Tennessee and Oak Ridge National Laboratory CUDA a architecture designed to facilitate development parallel programs. In conjunction comprehensive software platform, Architecture enables programmers draw on immense power graphics processing units (GPUs) when building high-performance applications. GPUs, course, have long been available demanding game now brings this valuable resource applications in other domains, including science, engineering, finance. No knowledge programming requiredjust ability program modestly extended version C. Example, written two senior members platform team, shows how employ new technology. The authors introduce each area through examples. After concise introduction architecture, as well quick-start guide C, details techniques trade-offs associated key feature. Youll discover use C extension write that delivers truly outstanding performance. Major topics covered include Parallel Thread cooperation Constant memory events Texture Graphics interoperability Atomics Streams multiple GPUs Advanced atomics Additional resources All tools youll need are freely download from NVIDIA.http://developer.nvidia.com/object/cuda-by-example.html