ARS-111GL-DNHR-LCC Supermicro GPU System

Supermicro GPU System ARS-111GL-DNHR-LCC - 1U - 2 nodes - NVIDIA GH200 Grace Hopper Superchip - up to 480GB memory - 4x E1.S NVMe - onboard H100 GPU - Liquid Cooling - 2700W Redundant

ARS-111GL-DNHR-LCC Supermicro GPU server has two nodes in a 1U form factor with liquide cooling. Each node is equipped with NVIDIA GH200 Grace Hopper Superchip and H100 GPU, supports up to 480GB memory and 4 hot-plug NVMe drives. It features IPMI 2.0 for remote management. For a comprehensive overview of advanced technologies, including the ARS-111GL-DNHR-LCC, visit our SERVERS solutions page.

Key Applications

  • High Performance Computing

  • AI/Deep Learning Training and Inference

  • Large Language Model (LLM) and Generative AI

Key Features

  • High density 1U 2-node GPU system with Integrated NVIDIA® H100 GPU (1per node) and liquid cooling

  • NVIDIA Grace Hopper™ Superchip (Grace CPU and H100 GPU)

  • NVLink® Chip-2-Chip (C2C) high-bandwidth, low-latency interconnect between CPU and GPU at 900GB/s

  • Up to 576GB of coherent memory per node including 480GB LPDDR5X and 96GB of HBM3 for LLM applications

  • 2x PCIe 5.0 x16 slots supporting NVIDIA BlueField®-3 or ConnectX®-7

  • 7 Hot-Swap Heavy Duty Fans with Optimal Fan Speed Control

This model is certified for Red Hat Enterprise Linux.

Discover more within the GPU servers product family.

To maintain quality and integrity, this product is sold only as a completely-assembled system (1 drive and 1 nework card per node installed). For more information about specialized system optimization, please contact Server Simply or visit the manufacturer's website.

Pending a firmware fix, this system currently supports two E1.S drives only per node. Please contact Server Simply for additional information about specialized system optimization.

All 2 nodes will be configured as selected in the current product configurator below.