Accelerate Time to Market for Generative AI and Beyond

Construct tailored AI solutions with Supermicro NVIDIA MGX™ Systems, featuring the latest NVIDIA GH200 Grace Hopper™ Superchip and NVIDIA Grace™ CPU Superchip. The new modular architecture is designed to standardize AI infrastructure and accelerated computing in compact 1U and 2U form factors while providing ultimate flexibility and expansion ability for the present and future GPUs, DPUs, and CPUs.

Supermicro’s advanced liquid-cooling technology enables power-efficient hyper-dense configurations, such as a 1U 2-node system with 2 NVIDIA GH200 Grace Hopper Superchips, each comprising of a single NVIDIA H100 Tensor Core GPU and an NVIDIA Grace CPU, integrated with a high-speed interconnect. Supermicro delivers thousands of rack-scale AI clusters per month from facilities around the world, ensuring plug and play compatibility.

Supermicro 1U NVIDIA Grace CPU Superchip system
Supermicro 1U Liquid-Cooled Grace Hopper MGX System (ARS-111GL-DNHR-LCC)

1U Grace Hopper MGX Systems

CPU+GPU Coherent Memory System for AI and HPC Applications

  • Up to 2 NVIDIA GH200 Grace Hopper Superchips featuring 72-core ARM CPU and H100 Tensor Core GPU tightly coupled with coherent memory
  • Up to 96GB HBM3 and 480GB LPDDR5X integrated memory per Grace Hopper Superchip
  • NVLink® Chip-2-Chip (C2C) high-bandwidth and low-latency interconnect
  • Up to 2 PCIe 5.0 x16 slots supporting NVIDIA BlueField®-3, NVIDIA ConnectX®-7 or additional GPUs
  • 8 hot-swap E1.S and 2 M.2 slots
  • Air Cooling and liquid cooling options
1U Grace Hopper MGX Systems Configurations at a Glance

Construct new solutions for accelerated infrastructures enabling scientists and engineers to focus on solving the world’s most important problems with larger datasets, more complex models, and new generative AI workloads. Within the same 1U chassis, Supermicro’s dual NVIDIA GH200 Grace Hopper Superchip systems deliver the highest level of performance for any application on the CUDA platform with substantial speedups for Al workloads with high memory requirements. In addition to hosting up to 2 onboard H100 GPUs in 1U form factor, its modular bays enable full-size PCIe expansions for present and future of accelerated computing components, high-speed scale-out and clustering.

Supermicro 2U Grace/x86 MGX System

Grace and x86 MGX Systems

Modular Building Block Platform Supporting today’s and Future GPUs, CPUs, and DPUs

  • NVIDIA Grace™ CPU Superchip (144-core) or 4th Gen Intel® Xeon® Scalable processor
  • Up to 480GB integrated LPDDR5X DRAM memory with ECC and up to 1TB/s of bandwidth (with Grace CPU Superchip) or up to 2TB 4800MT/s ECC DDR5 DRAM (with Intel CPU)
  • Up to 4 NVIDIA double-width PCIe GPUs, including H100 PCIe, H100 NVL, L40S, and more
  • Up to 8 Hot-Swap E1.S and 2 M.2 NVMe slots
  • Up to 3 PCIe 5.0 x16 slots supporting NVIDIA BlueField-3 or NVIDIA ConnectX-7
Grace and x86 MGX System Configurations at a Glance

Supermicro NVIDIA MGX™ 1U/2U Systems with Grace™ CPU Superchip and x86 CPUs are fully optimized to support up to 4 GPUs via PCIe without sacrificing I/O networking, or thermals. The ultimate building block architecture allows you to tailor these systems optimized for a variety of accelerated workloads and fields, including Al training and inference, HPC, data analytics, visualization/Omniverse™, and hyperscale cloud applications.

Resources