Hardware Datasheet · GPU Server
NVIDIA
NVIDIA DGX H100
Eight H100 SXM5 GPUs and 640GB HBM3 — the Hopper-generation DGX workhorse.
Overview
DGX H100 is the Hopper-generation NVIDIA system that a large share of production AI infrastructure was built on: eight H100 SXM5 GPUs, 640GB of HBM3, NVSwitch between them, and eight ConnectX-7 adapters for the cluster fabric. It remains a rational purchase for teams extending an existing DGX estate, where matching what you already run matters more than the newest silicon. Supplied configured, with the fabric and storage quoted alongside.
Specifications
| GPU / Accelerator | 8× NVIDIA H100 80GB SXM5 |
| GPU Memory | 640GB HBM3 total |
| GPU Interconnect | NVLink with NVSwitch, 900GB/s GPU-to-GPU |
| CPU | Dual Intel Xeon Platinum 8480C |
| System Memory | 2TB DDR5 |
| Storage | NVMe (capacity configurable; confirmed at quote) |
| Networking | 8× ConnectX-7 400Gb/s InfiniBand; 2× BlueField-3 DPU |
| Form Factor | 8U rackmount |
Typical Use Cases
- ·Large language model training at cluster scale
- ·Production inference on established Hopper workloads
- ·Adding capacity to an existing DGX H100 deployment
- ·SuperPOD-architecture scale units
Industries
Warranty & Support
Supplied through authorized channels with full manufacturer warranty. On-site, next-business-day support options available. Every system is configured, tested, and documented before delivery, with asset and warranty records provided for enterprise audit requirements.
Request a tailored quote
Configurations are tailored per engagement — contact us for pricing and lead times.
sales@nexus-compute.com
+1 737 276 1016
nexus-compute.com
Specifications are indicative and configured to each engagement. All product names, logos, and trademarks are the property of their respective owners. Nexus Compute is an independent enterprise hardware supplier.