GTC 2020: Under the Hood of the new DGX A100 System Architecture
After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
Under the Hood of the new DGX A100 System Architecture
Alex Ishii, NVIDIA | Nick Whidden, NVIDIA
Deep dive into the new DGX A100, NVIDIA's new Deep Learning Supercomputer with 8x A100 GPUs (Based on NVIDIA Ampere microarchitecture). Presenting system hardware architecture, feeds and speeds, new features and capabilities. Key performance metrics, and a summary of observed performance on DL/mL/HPC benchmark applications. Basic familiarity with server architectures, server components, and application-performance metrics assumed, and some familiarity with previous NVIDIA DGX server products a plus.