As an integral component of the adaptable NVIDIA MGX platform, the NVIDIA MGX GH200 delivers exceptional flexibility.
Through this modular framework, OEM and ODM partners can design targeted solutions for specific needs while optimizing R&D resources and speeding up deployment.
The architecture seamlessly accommodates diverse setups of GPUs, CPUs, and DPUs including Arm®, x86, NVIDIA Grace, and NVIDIA OVX™ systems to drive high-performance enterprise data center workloads.
Featuring a unified modular layout, the NVIDIA MGX architecture enables hundreds of distinct setup combinations on the same underlying platform.
NVIDIA MGX systems are compatible with a broad range of hardware, supporting x86 AMD and Intel platforms, ARM CPUs, and the entire lineup of current-generation NVIDIA GPUs.
Designed for future-proof integration, the MGX system’s modular layout seamlessly supports next-generation hardware and accelerates time to market.
NVIDIA MGX GH200 System is powered by the Hopper architecture, used in the H200 GPU.
| Format | High-density 1U GPU server |
| CPU | 72-core Grace Arm Neoverse V2 CPU + H100 Tensor Core GPU in a single chip |
| GPU | NVIDIA H100 Tensor Core GPU with 96GB of HBM3 or 144GB of HBM3e |
| Memory | 480GB integrated LPDDR5X with ECC |
| Drives | 8x hot-swap E1.S drives and 2x M.2 NVMe drives |
| Networking | 3x PCle 5.0 x16 slots supporting NVIDIA BlueField-3 or ConnectX-7 |
| Interconnect | NVLink-C2C with 900GB/s for CPU-GPU interconnect |
| Cooling | Air-cooling, 8 hot-swap high-flow fans |
| Power | 2x 2000W Redundant Titanium Level power supplies |
| CPU | 72-core Grace Arm Neoverse V2 CPU |
|---|---|
| GPU | NVIDIA H100 Tensor Core GPU with 96GB of HBM3 or 144GB of HBM3e |
| Memory | 480GB integrated ECC LPDDR5X |
| Drives | 8x hot-swap E1.S and 2x M.2 NVMe 960GB |
| Interconnect | NVLink-C2C with 900GB/s |
Your email address will not be published. Required fields are marked *
Reviews
There are no reviews yet.