Parallel Programming Session 5.2
Welcome to Session 5.2 of my Parallel Programming course ! In this video, we continue our deep dive into HIP — the Heterogeneous-Compute Interface for Portability — focusing on memory allocations, access patterns, and the unified memory model that enable efficient GPU programming. You will learn how to manage host and device memory explicitly and with unified memory, understand the GPU memory hierarchy, and optimize memory transfers to maximize performance.
We’ll also explore kernel optimizations and profiling techniques to help you write faster and more efficient GPU code. Plus, we’ll cover advanced topics such as shared memory, thread synchronization within kernels, and programming across multiple GPUs including how HIP works with MPI to leverage multi-node, multi-GPU parallelism.
With practical examples and explanations, this session will equip you with the tools and concepts needed to harness the full power of HIP for high-performance, portable parallel computing.
Don’t forget to like, subscribe, and stay tuned for the next sessions.
Sur cette page du site, vous pouvez voir la vidéo en ligne Parallel Programming Session 5.2 durée heure minute seconde en bonne qualité , qui a été Téléchargé par l'utilisateur Patrick Lemoine 07 août 2025, Partagez le lien avec vos amis et connaissances, sur youtube cette vidéo a déjà été regardée 61 fois et il a aimé 1 téléspectateurs. Bon visionnage!
