Scalable Infrastructure for AI Leadership
The NVIDIA DGX SuperPOD™ with NVIDIA DGX™ A100 systems is the most advanced next generation AI supercomputing infrastructure, providing groundbreaking computational power to solve the world’s most challenging computational problems and neural network models. It is an optimized system for multi-node mixed AI workload including training, development, inference, analytics and HPC. DGX SuperPOD is the only right tool for Enterprise AI Leaders.
NVIDIA DGX A100 System
DGX SuperPOD is configured to be a scalable and balanced system with the full 140 DGX A100 systems with a total of 1120 NVIDIA A100 GPUs and scaling to hundreds of nodes. The DGX SuperPOD design includes Mellanox networking switches, DGX POD certified storage, and NVIDIA GPU CLOUD® (NGC) optimized software.