SYS-422GA-NBRT-LCC

DP Intel 4U Liquid-Cooled Server with NVIDIA HGX B200 8-GPU

Anewtech-Systems-GPU-Server-Supermicro-SYS-422GA-NBRT-LCC Liquid-Cooled Supermicro GPU Server with NVIDIA HGX B200
  • Dual Intel® Xeon® 6900 series processors with P-cores, up to 500W TDP, 6UPI
  • Support for up to 8 onboard SXM GPU accelerator cards
  • 24 DIMM slots, ECC DDR5 designed for up to 6400MT/s (1DPC only)
  • 2 10G NIC (X710), 1 VGA, 2 USB 3.0, and 1 Dedicated IPMI
  • 8 PCIe 5.0 x16 LP and 2 PCIe 5.0 x16 FHHL
  • Up to 8 front hot-swap 2.5" NVMe drive bays and 2 NVMe M.2

Key Applications

  • High Performance Computing
  • AI/Deep Learning Training
  • Industrial Automation, Retail
  • Healthcare
  • Conversational AI
  • Business Intelligence & Analytics
  • Drug Discovery
  • Climate and Weather Modeling
  • Finance & Economics
Product Specification
Anewtech-Systems-GPU-Server-Supermicro-SYS-422GA-NBRT-LCC-Liquid-Cooled Supermicro GPU Server with NVIDIA HGX B200  Intel 4U Liquid-Cooled Server
Anewtech-Systems-GPU-Server-Supermicro-SYS-422GA-NBRT-LCC-Liquid-Cooled Supermicro GPU Server with NVIDIA HGX B200
Product SKUsSuperServer SYS-422GA-NBRT-LCC
MotherboardSuper X14DBG-DAP
Processor
CPUDual Socket BR (LGA-7529)
Intel® Xeon® 6900 series processors with P-cores
Core CountUp to 128C/256T; Up to 504MB Cache per CPU
NoteSupports up to 500W TDP CPUs (Liquid Cooled)​
GPU
Max GPU CountUp to 8 onboard GPUs
Supported GPUNVIDIA SXM: HGX B200 8-GPU (180GB)
CPU-GPU InterconnectPCIe 5.0 x16 CPU-to-GPU Interconnect
GPU-GPU InterconnectNVIDIA® NVLink® with NVSwitch™
System Memory
MemorySlot Count: 24 DIMM slots
Max Memory (1DPC): Up to 6TB 6400MT/s ECC DDR5 RDIMM 
Max Memory (1DPC): Up to 6TB 8800MT/s ECC DDR5 MRDIMM
On-Board Devices
ChipsetSystem on Chip
Network Connectivity2 RJ45 10GbE with Intel® X710-AT2
Input / Output
Video1 VGA port
System BIOS
BIOS TypeAMI 32MB SPI Flash EEPROM
Management
SoftwareSuperCloud Composer
Supermicro Server Manager (SSM)
Super Diagnostics Offline (SDO)
Supermicro Thin-Agent Service (TAS)
SuperServer Automation Assistant (SAA) New!
Power configurationsPower-on mode for AC power recovery
ACPI Power Management
Security
HardwareTrusted Platform Module (TPM) 2.0
Silicon Root of Trust (RoT) – NIST 800-193 Compliant
FeaturesCryptographically Signed Firmware
Secure Boot
Secure Firmware Updates
Automatic Firmware Recovery
Supply Chain Security: Remote Attestation
Runtime BMC Protections
System Lockdown
PC Health Monitoring
CPUMonitors for CPU Cores, Chipset Voltages, Memory
8+4 Phase-switching voltage regulator
FANFans with tachometer monitoring
Status monitor for speed control
Pulse Width Modulated (PWM) fan connectors
TemperatureMonitoring for CPU and chassis environment
Thermal Control for fan connectors
Chassis
Form Factor4U Rackmount
ModelCSE-GP403TS-R000NP
Dimensions and Weight
Height13.8" (174 mm)
Width17.7" (449 mm)
Depth34.62" (879.4 mm)
Package13" (H) x 48" (W) x 26.4" (D)
WeightN/A
Available ColorSilver
Front Panel
LEDHDD activity
System information (overheat/UID) 
UID
ButtonsPower On/Off 
UID/ Reset button
Expansion Slots
PCI-Express (PCIe) ConfigurationDefault
8 PCIe 5.0 x16 LP slots 
2 PCIe 5.0 x16 FHHL slots 
Option A
8 PCIe 4.0 x16 LP slots 
4 PCIe 5.0 x16 FHHL slots
Drive Bays / Storage
Drive Bays ConfigurationDefault: Total 8 bays
8 front hot-swap 2.5" NVMe drive bays
M.22 M.2 NVMe slots (M-key)
System Cooling
Fans4 Fan 8cm Fan(s)
Liquid CoolingDirect to Chip (D2C) Cold Plate (optional)
Power Supply
Power Supply4x 5250W Redundant Titanium (certification pending) Level (96%) power supplies
Dimension (WxHxL)106.5 x 82.1 x 245.3 mm
AC Input5250W: 200-240Vac / 50-60Hz
+12VMax: 125A / Min: 0A (200Vac-240Vac)
Output TypeBackplanes (gold finger)
Operating Environment
Environmental Spec.Operating Temperature: 10°C ~ 35°C (50°F ~ 95°F)
Non-operating Temperature: -40°C to 60°C (-40°F to 140°F)
Operating Relative Humidity: 8% to 90% (non-condensing)
Non-operating Relative Humidity: 5% to 95% (non-condensing)
 

X14 GPU-Optimized: Maximum acceleration for AI Training, LLMs, and Generative AI

Next-generation architecture for the most intensive AI workloads. Dual Intel® Xeon® 6900 series processors with P-cores.

 

Anewtech-Systems-Supermicro-GPU-Servers-SYS-A22GA-NBRT-Nvidia-GPU-Server.

SYS-A22GA-NBRT 
10U 8-GPU Air Cooled
NVIDIA SXM: HGX B200 8-GPU (180GB)
10 2.5” NVMe

Anewtech-Systems-Supermicro-GPU-Server-SYS-822GA-NBRT-Nvidia-GPU-Servers

SYS-822GA-NBRT 
8U 8-GPU Air Cooled
NVIDIA SXM: HGX H100 8-GPU (80GB),
 HGX H200 8-GPU (141GB)
10 2.5” NVMe

Anewtech-Systems-Supermicro-GPU-Server-SYS-422GA-NBRT-LCC-Supermicro-Liquid-Cooled Supermicro GPU Server with NVIDIA HGX B200

SYS-422GA-NBRT-LCC 
4U 8-GPU Liquid Cooled
NVIDIA SXM: HGX B200 8-GPU (180GB)
8 2.5” NVMe

4U Liquid-Cooled Rack Configuration

Scaling the infrastructure for multi-trillion parameter AI models, Supermicro is at the forefront of adopting networking innovations for both InfiniBand and Ethernet, including NVIDIA BlueField®-3 SuperNIC and ConnectX-7 at 400Gb/s, as well as ConnectX-8, Spectrum-4, and Quantum-3 to enable 800Gb/s networking for the NVIDIA Blackwell Platform.

The NVIDIA Quantum-2 InfiniBand and Spectrum-X Ethernet with Supermicro’s 4U liquid-cooled and 8U air-cooled NVIDIA HGX H100 and H200 system clusters now power one of the largest AI deployments to date.

4U 8-GPU System

SYS-422GA-NBRT-LCC / AS-4126GS-NBR-LCC

 

Overview4U Liquid-cooled System with NVIDIA HGX B200 8-GPU
CPUDual Intel® Xeon® 6 or AMD EPYC™ 9005 Series Processors
Memory• Up to 6TB DDR5-6400 or MRDIMM 8800MT/s (Intel)
• Up to 9TB DDR5-6000 (AMD)
GPUNVIDIA HGX B200 8-GPU (180GB HBM3e per GPU)
1.8TB/s NVLink GPU-GPU interconnect with NVSwitch
Networking8x NVIDIA ConnectX®-7 or BlueField®-3 SuperNICs
2x NVIDIA ConnectX®-7 Dual-port 200Gbps/NDR200 QSFP112 NICs
1:1 networking to each GPU to enable NVIDIA GPUDirect RDMA and Storage
Storage8 front hot-swap 2.5” PCIe 5.0 x4 NVMe drive bays
Power Supply4x 6.6kW redundant Titanium Level power supplies
Anewtech-Systems-GPU-Server-Supermicro-SYS-422GA-NBRT-LCC-liquid-cooled-server