Harnessing the Power of GPU-Accelerated Workflows with Alpine (A Primer)

CURC's largest supercomputer, Alpine, features state-of-the-art AMD and NVIDIA GPUs capable of dramatically slashing computation times for complex deep-learning workflows. However, transitioning from CPUs to GPUs often comes with a steep learning curve. This Primer* is specifically designed to help researchers with little to no GPU programming experience bridge the gap. This workshop introduces heterogeneous (CPU/GPU) computing, popular programming models, and best practices for porting, testing, and optimizing GPU-accelerated workflows. Participants will get hands-on experience requesting Alpine’s GPU resources via Slurm and using system tools to monitor real-time GPU usage and efficiency Participants should have a personal computer and a Research Computing Account prior to the start of the training if they wish to try hands-on examples. RMACC members should refer to our RMACC Access to Alpine page. *What is a Primer? A Primer is a session that provides an introduction to concepts, sy...