Introduction to GPU Programming with CUDA
The Luxembourg SuperComputing Competence Center will be hosting a half-day online introduction to programming with GPUs using CUDA on 30 April 2024. The course will be divided into two parts, with the first part focusing on theory and the second part dedicated to hands-on challenges using the GPU accelerators of the MeluXina supercomputer. To register for this event, please follow this link.
This course's participants will develop their GPU programming skills using the CUDA programming model. This will include learning about synchronization, memory allocation, and device and host calls. They will also gain an understanding of GPU architecture and how parallel thread blocks are used to parallelize computational tasks. Since the GPU is an accelerator, it is important to understand memory management between the GPU and CPU, which will also be discussed in detail. Additionally, participants will learn how to use the CUDA programming model to accelerate linear algebra routines and iterative solvers on the GPU. The course will begin with a theoretical introduction, followed by hands-on tutorials with mentor guidance.
Prospective and current users of large hybrid CPU/GPU clusters and supercomputers are encouraged to participate.
