SupermicroGPU & AI Servers

Supermicro 4U Liquid-Cooled HGX B200 System

Model SYS-422GS-NBRT-LCC

Front-I/O DLC-2 system enabling up to 96 Blackwell GPUs in a single rack.

  • GPUsNVIDIA HGX B200 8-GPU
  • DensityUp to 96 GPUs per rack
  • CoolingDLC-2 direct liquid cooling
  • Form factor4U, front I/O
Illustration of an 8-GPU AI server
Overview

About the 4U Liquid-Cooled HGX B200 System

This 4U liquid-cooled Supermicro server runs an NVIDIA HGX B200 8-GPU board with direct-to-chip cold plates. Eight systems with 64 GPUs fit in a 42U rack, or twelve systems with 96 GPUs in a 52U rack.

Supermicro's direct liquid cooling captures most system heat at the source, cutting data center power by up to 40% and running at noise levels as low as 50 dB.

HGX B200 8-GPU

1.44 TB of GPU memory with NVLink and NVSwitch.

Direct liquid cooling

Cold plates on CPUs and GPUs; up to 40% data center power savings.

Up to 96 GPUs per rack

12 systems in a 52U rack, or 64 GPUs in 42U.

Dual Intel Xeon 6700

Up to 86 cores and 350 W per CPU.

10 PCIe 5.0 slots

8× x16 LP for networking plus 2× x16 FHHL.

Redundant power

4× 6.6 kW Titanium (2+2).

Specifications

Technical Specifications

Processor & memory

CPUsDual Intel Xeon 6700 series with P-cores, up to 86C/172T, 336 MB cache, up to 350 W
Memory32 DIMM slots; up to 4 TB DDR5-6400 (1DPC) or 8 TB DDR5-6000 (2DPC)

GPU

GPUsNVIDIA HGX B200 8-GPU (180 GB each)
GPU memory1.44 TB per system
InterconnectNVIDIA NVLink with NVSwitch

Storage & expansion

Drive bays8× front hot-swap E1.S NVMe
Boot2× M.2 NVMe with RAID
Expansion8× PCIe 5.0 x16 LP, 2× PCIe 5.0 x16 FHHL
LAN2× 10 GbE RJ45 (Intel X710-AT2)

Chassis & power

Form factor4U rackmount, front I/O
Dimensions (H × W × D)174 × 448 × 991 mm
Weight85 kg net
Power4× 6.6 kW redundant (2+2) Titanium (96%)
CoolingDirect-to-chip cold plates, 4× 80 mm fans
SecurityTPM 2.0, Silicon Root of Trust, Secure Boot, remote attestation
Source: official Supermicro product information. Specifications are set by the manufacturer and may change. Plan Man confirms the final configuration, availability and lead time in your quote.
Use cases

What it's built for

  • AI and deep learning training
  • LLM and multimodal model development
  • Inference at scale
  • HPC
Plan Man services

Supplied and supported in the UAE

  • Requirement analysis, sizing and bill of materials
  • Genuine supply with manufacturer warranty
  • Rack integration, cabling and onsite deployment
  • Power and liquid cooling planning for dense GPU racks
  • After-sales support, RMA handling and upgrades from Dubai

Get a Quote

Tell us your workload and timeline for the Supermicro 4U Liquid-Cooled HGX B200 System. We reply within one business day with a configuration, pricing and lead time.

Request a Quote