Key Applications

  • Scientific Research
  • Conversational AI
  • Business Intelligence & Analytics
  • Drug Discovery
  • Finance Services and Fraud Detection
  • AI/Deep Learning Training and Inference
  • Large Language Model (LLM) and Generative AI
  • High Performance Computing (HPC)
  • Autonomous Vehicle Technologies

Key Features

  1. Dual AMD EPYC™ 9005/9004 Series Processors
  2. NVIDIA HGX B300 8-GPU with 5th Generation NVLink® 1.8TB/s, 2.3TB of HBM3e GPU memory per system
    • 24 DIMMs, with up to 6TB at 6400MT/s DDR5 memory
    • (Please check System Memory Section for detail)
  3. 8 Hot-swap E1.S NVMe storage drive bays and 2 M.2 NVMe boot drives
  4. 8 NVIDIA ConnectX®-8 SuperNICs, up to 800Gb/s, up to two dual-port NVIDIA BlueField®-3 DPUs
    • 6x 6600W Redundant (3 + 3) Titanium Level (96%) power supplies
    • (Power supply full redundancy based on configuration and application load)

NCS International

Den Sliem 89
7141 JG Groenlo
The Netherlands
+31 544 470 000
info@ncs.nl

The Blueprint for Modern AI Data Centers

Supermicro’s Data Center Building Block Solutions (DCBBS) delivers complete, modular AI infrastructure. Built from validated components and sub-systems, DCBBS provides end-to-end deployment flexibility – from individual GPUs and networking switches to complete racks, site infrastructure, management software, and professional services.

Liquid-Cooled Universal GPU Systems

Direct-to-chip liquid-cooled systems for high-density AI infrastructure at scale.
  • GPU:NVIDIA HGX H100/H200/B200
  • CPU:Intel® Xeon® or AMD EPYC™
  • Memory:Up to 32 DIMMs, 9TB
  • Drives:Up to 24 Hot-swap U.2 or 2.5″ NVMe/SATA drives

Microcloud 3U 5 Nodes

CPU + GPU Multi-Node System for AI Driving Workloads
  • CPU:Single AMD EPYC™ 4005/4004 series processor
  • GPU:Up to 1 double-width PCIe GPU per node
  • Memory:Up to 4 DIMMs, 192GB DDR5
  • Drives:Up to 2 internal fixed 2.5″ PCIe 5.0 NVMe + 2 internal fixed 2.5″ SATA per node
  • PCIe:Up to 1 PCIe 5.0 x16 FHFL slot + 1 PCIe 5.0 x4 MicroLP slot per node

X14 Petascale All-Flash

The future for Low-Latency, High-Throughput Storage
Up to 1.92PB (PetaByte) of flash storage in 2U

  • Dual socket Intel® Xeon® 6700/6500 series processors with
    P-cores or 6700 series processors with E-cores
  • 32 DIMM slots per node supporting DDR5-6400 and support
    for MRDIMMs up to 8000MT/s (P-core only)
  • Up to 2 PCIe Gen 5 AIOM slots and 3 PCIe Gen 5 slots
  • Support for the latest industry standard EDSFF E3.S drive
    form factors and Gen5 drives from all major SSD vendors
  • Optional CXL support to provide up to 2TB of additional
    memory
  • Optional double-width GPU support up to 300W TDP
  • Engineered to minimize latency and maximize airflow over
    critical components

Reaching New Heights of Density and Power Efficiency

Supermicro’s all-new redesigned H14 generation solutions offer unparalleled performance and flexibility for a wide range of AI and HPC applications. Building on Supermicro’s proven building-block architecture, these solutions help enterprise customers efficiently upgrade and scale their workloads. Supermicro offers one of the broadest portfolios of AMD EPYC servers, featuring the latest EPYC 9005 series processors to provide maximum core count, density, and efficiency.

The H14 generation servers feature the latest leading GPUs including the new AMD Instinct MI350 Series and upcoming NVIDIA HGX B200 GPUs. The new H14 generation servers with the combination of the latest EPYC CPU and Instinct accelerators truly offers a powerhouse of AI-enabled solutions to market.