GPU programming Part 2: Advanced GPU Programming

GPU-accelerated computing is driving current scientific research. Writing fast numerical algorithms for GPUs provides high application performance by offloading computationally intensive parts of the code to a GPU.

The course covers aspects of GPU architectures and programming. Emphasis is placed on the use of the CUDA C++ parallel programming language, which allows maximum control over NVIDIA GPU hardware.

Examples of increasing complexity are used to demonstrate optimization and tuning of scientific applications. This advanced course consists of modules that cover multi-GPU programming, modern CUDA concepts, CUDA Fortran, and portable programming models such as OpenACC and parallel STL algorithms in C++ in more detail.

Target group: Advanced programmers, programming scientists, students (max. 30 participants)

Language: English

Application deadline: 5.6.