This title appears in the Scientific Report :
2023
Please use the identifier:
http://dx.doi.org/10.34734/FZJ-2023-05224 in citations.
GPU Programming Part 2: Advanced GPU Programming
GPU Programming Part 2: Advanced GPU Programming
GPU-accelerated computing drives current scientific research. Writing fast numeric algorithms for GPUs offers high application performance by offloading compute-intensive portions of the code to a GPU.The course covers aspects of GPU architectures and programming. Focus is on the usage of the parall...
Saved in:
Personal Name(s): | Meinke, Jan |
---|---|
Herten, Andreas (Corresponding author) / Hrywniak, Markus / Kraus, Jiri / Haghighi Mood, Kaveh | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
2023
|
DOI: |
10.34734/FZJ-2023-05224 |
Conference: | JSC - as part of the Training Programme of Forschungszentrum Jülich, online (Germany), 2023-06-19 - 2023-06-23 |
Document Type: |
Lecture |
Research Program: |
Future Computing & Big Data Systems Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups |
Link: |
Get full text OpenAccess OpenAccess OpenAccess OpenAccess OpenAccess OpenAccess OpenAccess OpenAccess OpenAccess |
Publikationsportal JuSER |
GPU-accelerated computing drives current scientific research. Writing fast numeric algorithms for GPUs offers high application performance by offloading compute-intensive portions of the code to a GPU.The course covers aspects of GPU architectures and programming. Focus is on the usage of the parallel programming language CUDA C++, which allows maximum control of NVIDIA GPU hardware. Examples of increasing complexity are used to demonstrate optimization and tuning of scientific applications. The foundations of GPU programming are covered in another dedicated Basic Course.This advanced course consists of modules providing more in-depth coverage of multi-GPU programming, modern CUDA concepts, CUDA Fortran, and portable programming models such as OpenACC and C++ parallel STL algorithms. |