Tuesday, November 19, 2019

SC19: NVIDIA CEO Jensen Huang on the Expanding Universe of HPC


NVIDIA CEO Jensen Huang addresses 1,400+ attendees of SC19, the annual supercomputing conference, in Denver, where he introduced a reference design for building GPU-accelerated Arm servers, announced the world̢۪s largest GPU-accelerated cloud-based supercomputer on Microsoft Azure, and unveiled NVIDIA Magnum IO storage software to eliminate data transfer bottlenecks for AI, data science, and HPC workloads. Jump ahead to: 05:10 - Lamborghini demo: Simulating computer graphics with NVIDIA RTX 10:11 - HPC milestones: Computation at petascale levels 17:36 - Accelerating the full stack of HPC 23:24 - HPC expanding in every direction 27:09 - AlexNet and BERT: Incredible advances in AI 40:07 - AI for science demo: Accelerating simulation 50:33 - Streaming AI 1:02:33 - Building supercomputers in the cloud 1:06:30 - Detecting neutrinos using cloud simulation 1:13:21 - World's largest on-demand supercomputer - NVIDIA GPUs on Azure cloud 1:21:22 - Introducing NVIDIA HPC reference platform for Arm-based servers 1:26:39 - VMD demo - molecular visualization tool 1:30:08 - Announcing NVIDIA Magnum IO 1:41:14 - Mars lander demo 1:47:47 - Wrapping up

No comments: