Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Lecture
Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture covers the advanced concepts of GPU computing, focusing on memory hierarchy, CUDA programming model, and data-parallel computing. It explains the GPU memory model, thread cooperation, and the CUDA programming model's intricacies. The lecture delves into shared memory, global memory, constant memory, and texture memory, highlighting their roles in optimizing GPU performance. It also discusses memory issues, coalescing, and bank conflicts, providing insights into efficient memory access strategies. Practical examples, such as matrix transpose optimization using shared memory, demonstrate the importance of memory management in achieving high performance in GPU computing.