Supermicro ARS-111GL-NHR 1U NVIDIA GH200 Grace Hopper MGX
Coherent Grace Hopper superchip computing in a 1U MGX node
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Configuration Options
Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.
NVIDIA GH200 Grace Hopper (integrated H100 GPU)
72-core NVIDIA Grace (Arm Neoverse V2)
96GB HBM3 + 480GB LPDDR5X (576GB total)
8x E1.S hot-swap NVMe, 2x M.2 NVMe
Overview
The ARS-111GL-NHR is a 1U NVIDIA MGX server built on a single GH200 Grace Hopper Superchip, pairing a 72-core Arm Grace CPU with an integrated H100 GPU over a 900GB/s NVLink-C2C link. It is specified, integrated, and validated to the target HPC or LLM profile, then supplied with manufacturer warranty through authorized channels.
Who This Solution Is For
Business Benefits
Coherent memory pool
576GB of unified CPU-GPU memory lets models and datasets exceed traditional HBM limits without manual staging.
1U scale-out density
The single-socket MGX node racks efficiently for large coherent clusters with minimal footprint.
Validated MGX build
Each superchip node is firmware-aligned and benchmarked before warranty-backed delivery.
Typical Business Use Cases
Large language model training and serving
Memory-bound HPC simulation
Graph analytics and vector search
Scientific data processing at scale
Industry Applications
Technical Overview
The platform is a single NVIDIA GH200 Grace Hopper Superchip MGX system uniting a 72-core Grace CPU and H100 GPU through a 900GB/s NVLink Chip-2-Chip coherent interconnect. It combines 96GB HBM3 with 480GB LPDDR5X for 576GB of addressable memory and supports NVIDIA BlueField-3 or ConnectX-7 across three PCIe 5.0 x16 slots.
| Accelerator | NVIDIA GH200 Grace Hopper (integrated H100 GPU) |
| Interconnect | NVLink-C2C, 900GB/s coherent CPU-GPU |
| CPU | 72-core NVIDIA Grace (Arm Neoverse V2) |
| Memory | 96GB HBM3 + 480GB LPDDR5X (576GB total) |
| Storage | 8x E1.S hot-swap NVMe, 2x M.2 NVMe |
| Networking/Fabric | BlueField-3 / ConnectX-7 ready, 3x PCIe 5.0 x16 |
| Form Factor | 1U rackmount |
| Power | Redundant 2000W power supplies |
| Management | ASPEED BMC, IPMI 2.0, Redfish |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
What advantage does Grace Hopper give over a PCIe H100?
The 900GB/s NVLink-C2C link makes 576GB of CPU and GPU memory coherent, so memory-bound LLM and HPC workloads avoid the PCIe bottleneck and CPU-GPU data copies.
Is software Arm-compatible?
Grace runs the Arm-native NVIDIA AI and HPC software stack; we validate the OS and CUDA environment for your workload before delivery.
How does it scale beyond one node?
BlueField-3 or ConnectX-7 adapters connect multiple 1U nodes into a high-bandwidth cluster fabric for distributed training and simulation.
Hardware Assistance
Configure the Supermicro ARS-111GL-NHR 1U NVIDIA GH200 Grace Hopper MGX with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.