2025年1月21日 · NVIDIA DGX SuperPOD: Next Generation Scalable Infrastructure for AI Leadership Reference Architecture Featuring NVDIA DGX B200

The NVIDIA DGX SuperPOD™ with NVIDIA DGX™ H100 system provides the computational power necessary to train today’s state-of-the-art deep learning (DL) models and to fuel …

The DGX SuperPOD architecture integrates NVIDIA software solutions including NVIDIA Base Command™, NVIDIA AI Enterprise, CUDA, and NVIDIA Magnum IO™. These technologies …

Organizations wanting to deploy their own supercomputing infrastructure should use the NVIDIA DGX SuperPOD Solution for Enterprise, which offers the DGX SuperPOD RA deployed in a …

Learn the secrets behind designing some of the world's highest-performing supercomputers. For a deeper dive into building similar clusters to supercharge your generative AI applications, watch …

NVIDIA DGX SuperPOD ™ 是一个 AI 数据中心基础设施平台，可让 IT 为每个用户和工作负载提供不打折扣的性能。在 DGX SuperPOD 的助力下，业内优秀的加速基础设施可以做到兼具敏捷 …

Using NVIDIA DGX-2H as its foundational compute building block, DGX SuperPOD integrates 96 nodes, combined in a fully connected low-latency network architecture using InfiniBand …

This paper describes key aspects of the DGX SuperPOD architecture including and how each of the components was selected to minimize bottlenecks throughout the system, resulting in the …

Leveraging the DGX SuperPOD designs reduces time-to-solution and time-to-market of next generation models and applications. The DGX SuperPOD is the integration of key NVIDIA …

All design is based on a radix optimized approach for Sharpv2 support and fabric performance and to align with design of Mellanox Quantum switches. DGX Pod: DGX Servers w. DGX base …