We take cryptocurrency — so the coins you hold can buy real hardware.Pay in crypto, spend it on real hardware. Free insured shipping worldwide over $399. Plain, unbranded boxes, sent from Germany.Free insured shipping over $399. Your rate is locked for 30 minutes once checkout opens.Rate locked 30 minutes. Every item sealed, serial-checked and covered by a 24-month warranty.Sealed, 24-month warranty.

Saved items Sign in
5U rackmount server chassis, matte black finish

GIGABYTE

GIGABYTE G593-SD2-AAX1 barebone server 32 DIMMs 8 GPU H100

SKU CH-3Y58-4QR92 MPN G593-SD2-AAX1

5U barebone server holding 32 DDR5 DIMM slots and supporting 8 NVIDIA HGX H100 GPUs with 900 GB/s GPU-to-GPU bandwidth

  • Dual Xeon processors handle massive AI workloads
  • 32 DDR5 DIMM slots enable huge memory capacity
  • NVIDIA HGX H100 platform with 8 SXM5 GPUs
  • 900GB/s GPU-to-GPU bandwidth via NVLink and NVSwitch
  • NVIDIA-Certified System validated for data centre deployment

HPC barebone for eight H100 GPUs

This 5U server barebone is built around the NVIDIA HGX H100 platform with eight SXM5 GPUs delivering 900GB/s GPU-to-GPU bandwidth via NVLink and NVSwitch. It targets organisations deploying large-scale AI training and high-performance computing workloads that demand maximum accelerator density in a single chassis.

Installation requires rack infrastructure

The unit arrives as a barebone system without processors, memory or GPUs pre-installed, so integration demands a compatible rack, adequate power distribution and cooling capacity for the 5U form factor. Common issues include underestimating the weight and airflow requirements of the fully populated HGX board, and neglecting the dual ROM architecture when planning firmware updates.

Suits AI clusters needing certified eight-GPU

Buyers who need a NVIDIA-Certified System with eight H100 GPUs, 32 DDR5 RDIMM slots and support for dual Xeon Scalable or Xeon CPU Max Series processors will match this platform. Teams requiring fewer GPUs, a smaller rack footprint, or a fully assembled appliance with validated software stacks should evaluate alternative configurations.

Highlights

  • Dual Xeon processors handle massive AI workloads
  • 32 DDR5 DIMM slots enable huge memory capacity
  • NVIDIA HGX H100 platform with 8 SXM5 GPUs
  • 900GB/s GPU-to-GPU bandwidth via NVLink and NVSwitch
  • NVIDIA-Certified System validated for data centre deployment

Specifications

No published specifications for this item yet.

Questions about this item

How can I verify the 900GB/s GPU-to-GPU bandwidth is active after setup?

Confirm the link speed in NVIDIA management tools and run a multi-GPU benchmark that saturates the NVLink fabric between the eight SXM5 modules.

Which major components are included and which must be sourced separately?

The barebones chassis, motherboard, NVLink interconnect and cooling for eight SXM5 GPUs are supplied; you must add two Xeon Scalable CPUs, 32 DDR5 RDIMMs and the H100 GPU modules.

Should I choose the 5th or 4th Gen Xeon Scalable platform for this build?

Both generations fit the dual-socket design; select 5th Gen for higher core counts and newer instruction support, or 4th Gen if existing licenses or validated stacks require it.

New to this? Buying server hardware with crypto — what to check, what it costs to pay, and what is never asked for.

Also in this aisle

People compared these

Same shelf, same checkout — eight coins and a 30-minute rate lock.