NH4230 is an 4U AI server based on Intel® 4th Gen Xeon® Scalable Processor. With its module design, its scalability, reliability, manageability and serviceability could fully support the AI business on both cloud datacenter and enterprise IT.

Key features
Performance Architecture
  • Support Two Intel® 4th Gen Xeon® Scalable Processor and 32DIMM DDR5 memories up to 4800MT/s
  • Direct connect topologic between CPU and GPU without retimer
  • Dual PCIe root connection between CPU to GPU in switch mode
Flexible Configuration
  • Up to 10x Dual width GPU Slots
  • Multiple CPU-GPU and GPU-GPU topologies
  • Single root and dual root connection between CPU and GPU in Switch mode
  • Up to 12 LFF or 24 SFF storage drives

Technical Specifications
Form Factor 4U Rack
Processor 2 x 4th Gen Intel ® Xeon ® scalable processors, TDP 350W
Memory Up to 32 DDR5 DIMMs and eight memory channels for each CPU
  • Up to 12 x 3.5-inch SATA/SAS/NVMe
  • Up to 24 x 2.5-inch SATA/SAS/NVMe
  • 1 x 2280 M.2 onboard
Expansion Slots
  • Rear PCIe Slot: 8 x FHFL DW GPU+5 x FHHL+1 x OCP
  • Front PCIe Slot: 2 x FHHL+1 x OCP/FHHL
  • AST2600
  • 2 x dedicated RJ45 MGMT port
Base IO Front
  • 1 x USB3.0+1 x USB2.0 Type-A Receptacle
  • 1 x PWR BTN with 1 x PWR Indicator
  • 1 x Status LED
  • 1 x VGA
  • 1 x UART Port Over TypeC
  • 1 x UID button & LED
  • 1 x GbE MGMT Port (Not both of Rear MGMT Port)
  • 1 x UID button& LED
  • 1 x GbE MGMT Port (Not both of Front MGMT Port)
  • 12 x 6056 FANs, N+1 Redundant
Power Supply 2+2 or 3+1 Redundant (2700W/3200W)
Dimension 175mm(H) x 447mm(W) x 900mm(L)
Operation Temperature 5ºC - 35ºC