欢迎大家赞助一杯啤酒🍺 我们准备了下酒菜:Formal mathematics/Isabelle/ML, Formal verification/Coq/Agda, C++/Erlang/Lisp
OpenACC
来自开放百科 - 灰狐
(版本间的差异)
小 |
小 (→图集) |
||
第8行: | 第8行: | ||
==指南== | ==指南== | ||
+ | |||
+ | ==文档== | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5382-OpenACC-2.5-and-Beyond.pdf OpenACC 2.5 and Beyond] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5195-Advanced-OpenACC-Programming.pdf Advanced OpenACC Programming] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5160-Experiences-in-Porting-Scientific-Applications-to-GPUs-Using-OpenACC.pdf Experiences in Porting Scientific Applications to GPUs Using OpenACC] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5711-Hands-on-Lab-Multi-GPU-Programming-with-MPI-and-OpenACC.pdf Hands on Lab Multi GPU Programming with MPI and OpenACC] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5233-GPU-Acceleration-Using-OpenACC-and-C++-Classes.pdf GPU Acceleration Using OpenACC and C++ Classes] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5192-Introduction-to-Compiler-Directives-with-OpenACC.pdf Introduction to Compiler Directives with OpenACC] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5196-Comparing-OpenACC-and-OpenMP-Performance-and-Programmability.pdf Comparing OpenACC and OpenMP Performance and Programmability] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5297-A-Simulation-of-Global-Atmosphere-Model-NICAM-on-TSUBAME2.5-Using-OpenACC.pdf A Simulation of Global Atmosphere Model NICAM on TSUBAME2.5 Using OpenACC] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5388-OpenACC-for-Fortran-Programmers.pdf OpenACC for Fortran Programmers] | ||
+ | *[http://docs.huihoo.com/gputechconf/gtc2015/S5139-Showing-the-Missing-Middle-Enabling-OpenACC-Performance-Analysis.pdf Showing the Missing Middle Enabling OpenACC Performance Analysis] | ||
==图集== | ==图集== |
2016年4月9日 (六) 15:16的版本
您可以在Wikipedia上了解到此条目的英文信息 OpenACC Thanks, Wikipedia. |
OpenACC 是一种全新的开放式并行编程标准,该标准旨在让数以百万计的科技类 Fortran 和 C 语言程序员能够轻松利用异构 CPU/GPU 计算系统的强大能力改变行业面貌。
OpenACC 让并行程序员能够为编译器提供简单的提示,亦称「指令」,使编译器能够识别哪些代码部分需要加速,无需程序员修改或改编底层代码本身。 在把计算任务映射到加速器上这方面,指令通过向编译器呈现出并行机制,从而让编译器能够执行这方面的详细工作。
OpenACC 指令由 PGI、Cray 以及英伟达在 CAPS 的支持下开发而成,是多家企业希望利用指令来简化 GPU 编程模型的一个共同愿景。这些企业均致力于支持一种共同的编程标准。
目录 |
指南
文档
- OpenACC 2.5 and Beyond
- Advanced OpenACC Programming
- Experiences in Porting Scientific Applications to GPUs Using OpenACC
- Hands on Lab Multi GPU Programming with MPI and OpenACC
- GPU Acceleration Using OpenACC and C++ Classes
- Introduction to Compiler Directives with OpenACC
- Comparing OpenACC and OpenMP Performance and Programmability
- A Simulation of Global Atmosphere Model NICAM on TSUBAME2.5 Using OpenACC
- OpenACC for Fortran Programmers
- Showing the Missing Middle Enabling OpenACC Performance Analysis
图集
链接
分享您的观点