NVIDIA DGX GH200

SKU DGX GH200-1 Category Brand:

  • 32x NVIDIA Grace Hopper Superchips, interconnected with NVIDIA NVLink
  • Massive, shared GPU memory space of 19.5TB
  • 900 gigabytes per second (GB/s) GPU-to-GPU bandwidth
  • 128 petaFLOPS of FP8 AI performance

About

The NVIDIA DGX GH200’s massive shared memory space uses NVLink interconnect technology with the NVLink Switch System to combine 256 GH200 superchips, allowing them to perform as a single GPU. This provides 1 exaflop of performance and 144 terabytes of shared memory — nearly 500x more memory than the previous generation NVIDIA DGX A100, which was introduced in 2020.

Specification

Specification
CPU and GPU
32x NVIDIA Grace Hopper Superchips
CPU Cores
2,304 Arm® Neoverse V2 Cores with SVE2 4X 128b
Shared Memory
19.5 TB
Performance
128 petaFLOPS of FP8 AI performance
Networking
32x OSFP single-port NVIDIA ConnectX-7 VPI with 400Gb/s InfiniBand 16x dual-port NVIDIA BlueField®-3 VPI with 200Gb/s InfiniBand and Ethernet
NVIDIA NVLink Switch System
9x L1 NVIDIA NVLink Switches
Management Network
Host baseboard management controller (BMC) with RJ45
Networking
4x OSFP ports serving 8x single-port NVIDIA ConnectX-7 400Gb/s InfiniBand/Ethernet 2x dual-port NVIDIA BlueField-3 DPUs VPI 1x 400Gb/s InfiniBand/Ethernet 1x 200Gb/s InfiniBand/Ethernet
Software
NVIDIA AI Enterprise (optimized AI software) NVIDIA Base Command (orchestration, scheduling, and cluster management) DGX OS / Ubuntu / Red Hat Enterprise Linux / Rocky (operating system)

Our Partners

Sign up with our newsletter to follow the latest trends in server technology