GPU并行计算编程技术介绍 |
| |
作者姓名: | 王泽寰 王鹏 |
| |
作者单位: | 英伟达公司,北京 100004 |
| |
摘 要: | 近年来GPU通用计算蓬勃发展。程序开发者和GPU通用计算应用程序的数量增长很快。针对不同的应用程序的要求和程序开发者不同的使用习惯,围绕着CUDA架构的 GPU,NVIDIA及其合作伙伴共同开发了很多种不同的编程技术。本文详细介绍了它们的特点和适用对象。希望可以帮助广大开发人员针对自己的编程习惯和程序要求选择最为合适的编程技术。
|
关 键 词: | GPU高性能计算 CUDA 编程技术 |
收稿时间: | 2012-11-20 |
Introduction to GPU Parallel Programming Technology |
| |
Authors: | Wang Zehuan Wang Peng |
| |
Affiliation: | NVIDIA Cooperation, Beijing 100004, China |
| |
Abstract: | Recently, GPGPU is developing rapidly. We can see a speedy growth in the number of programmers and the GPGPU applications. Tailoring to the demands of different applications and the various programming habits of the programmers, NVIDIA has co-developed different technologies used on GPU of CUDA architecture with its partners. This article is written for the purpose of making an elaborated introduction to them, their respective characteristics, and their target users. It is also written in the hope that programmers and developers can choose their most suitable technologies. |
| |
Keywords: | GPU High performance computing CUDA Programming technology |
|
| 点击此处可从《》浏览原始摘要信息 |
|
点击此处可从《》下载全文 |
|