The Advanced Programming of High-Performance GPU Architectures training course provides experienced students with advanced knowledge and hands-on experience in developing and analyzing high performance applications software for processors with massively parallel computing resources (graphics processing units and multicore processors).
The course begins with students learning algorithm styles that are suitable for accelerators. Next, students will gain an understanding of the most important architectural performance considerations for developing applications. The course concludes with students being exposed to computational thinking skills for accelerating applications in science and engineering and gaining the ability to engage computing accelerators on science and engineering breakthroughs.
The remote lab environment utilizes Nvidia hardware (Nvidia GTX480 and Tesla C2070) to illustrate CUDA/OpenCL concepts and to allow training participants to experimentally investigate performance issues, debugging techniques, and code examples.
Purpose
|
Learn about advanced programming techniques for OpenCL programming on Intel, AMD, and Nvidia architectures. |
Audience
|
Experienced programmers wanting to take a leadership role as a GPU project architect. |
Role
| Software Developer |
Skill Level
| Advanced |
Style
| Hack-a-thon |
Duration
| 1 Day |
Related Technologies
| CUDA |
Productivity Objectives
- Develop and analyze high performance application software for processors.
- Describe algorithm styles that are suitable for accelerators.
- Explain architectural performance for developing applications.
- Explore computational thinking for accelerating applications.