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 GPU memory hierarchy, including global, local, shared memory, and caches. It explains the challenges of SIMT execution, such as banked shared memory and inefficient global memory access. Optimization strategies are discussed, emphasizing coalescing memory accesses, reducing bank conflicts, eliminating warp divergence, and minimizing excess instructions.