NH8230 8U AI SERVER (NVIDIA HGX H100)
NH8230 is an 8U AI server based on Intel® 4th generation Xeon® Scalable Processor. With the NVIDIA HGX H100 module,it’s extremely optimized for hyper scale AI training and inference application for both cloud datacenter and enterprise IT.







Key features
Extreme AI Performance
  • Support Two Intel® 4th Gen Xeon® Scalable Processor and 32DIMM DDR5 memories up to 4800MT/s
  • Compatible with Intel® 5th Gen Xeon® Scalable Processor; DDR5 memory up to 5600 MT/s
  • Compatible with the latest NVIDIA HGX H100 module and HGX H200 module
Flexible System Architecture
  • Flexible x16 or x32 PCIe link bandwidth between CPU and Switch
  • Full height and half height NIC card available
  • SmartNIC supported



Technical Specifications
Form Factor 8U Rack
Processor 2 x 4th Gen Intel ® Xeon ® scalable processors, TDP 350W
GPU NVIDIA HGX™ H100 8GPU
Memory Up to 32 DDR5 DIMMs and eight memory channels for each CPU
Storage
  • Up to 8 NVMe SSDS (from Switch)
  • Up to 2 NVME/SATA SSDs (from CPU /PCH)
  • 1 onboard M.2 SSD (from PCH)
Expansion Slots
  • Up to 8 FHHL PCIe slots (from Switch)
  • Up to 4 FHHL PCIe slots (from CPU)
Management
  • BMC:AST2600
  • 1 RJ45 Gigabit management network port
Front IO
  • 2x USB, Type-A
  • 1x PWR BTN with LED
  • 1x UID BTN with LED
  • 1x Status LED
  • 1x Console PORT, Type C
  • 1x Front VGA
Cooling
  • 15 high-performance hot-swappable fans with N+1 redundancy
Power Supply
  • Up to 8 x 54V PSU (2 are optional) and 3+3 redundancy
  • 2 x 12V PSU and 1+1 redundancy
Dimension 448mm(H) *352mm(W) *900mm(L)
Operation Temperature 5ºC - 35ºC