ARMing GPUs for Impactful Science with the GH200 Superchip and True Exascale
Scalable Parallel Computing Lab, SPCL @ ETH Zurich via YouTube
Google Data Analytics, IBM AI & Meta Marketing — All in One Subscription
Become an AI & ML Engineer with Cal Poly EPaCE — IBM-Certified Training
Overview
Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Explore NVIDIA's GH200 Superchip architecture through a comprehensive technical analysis that examines whether this ARM-based system represents a breakthrough for HPC workloads or merely serves as an expensive memory controller. Dive deep into the complex memory system architecture of the GH200, uncovering technical intricacies and hidden features while learning practical strategies for leveraging this powerful hardware for large-scale scientific computing applications. Discover how to harness the GH200's capabilities for revolutionary computational science projects that push the boundaries of exascale computing, with insights from the Scalable Parallel Computing Lab at ETH Zurich on optimizing performance for impactful scientific research.
Syllabus
ARMing GPUs for Impactful Science with the GH200 Superchip (and true Exascale)
Taught by
Scalable Parallel Computing Lab, SPCL @ ETH Zurich