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
GIGABYTE G492-ZD0 4U barebone server chassis with eight GPU modules installed

GIGABYTE

GIGABYTE G492-ZD0 barebone server EPYC 7003 32 DIMMs 8-GPU

SKU CH-0RF9-EYZV0 MPN G492-ZD0

A 4U barebone server for AMD EPYC 7003 dual processors with 32 DDR4 DIMM slots and eight NVIDIA HGX A100 GPUs linked at 600 GB/s

  • NVIDIA-Certified System for data-centre reliability
  • Eight NVIDIA HGX A100 SXM4 GPUs for massive parallel compute
  • 600 GB/s GPU-to-GPU bandwidth via NVLink and NVSwitch
  • Dual AMD EPYC 7003 processors with 3D V-Cache technology
  • 32 DDR4 DIMM slots supporting eight-channel memory per CPU

Dual-socket EPYC server barebone

The GIGABYTE G492-ZD0 is a 4U server barebone built for high-performance computing and artificial-intelligence workloads. It accepts two AMD EPYC 7003 series processors, including models with AMD 3D V-Cache technology. The platform is designed for organisations that need dense GPU acceleration in a single chassis.

Eight SXM4 GPUs with NVLink and NVSwitch

The system integrates an NVIDIA HGX A100 baseboard with eight SXM4 GPUs connected by NVLink and NVSwitch fabric. This topology delivers up to 600GB/s GPU-to-GPU bandwidth without occupying PCIe slots. Thirty-two DDR4 DIMM slots across both processors provide eight memory channels per CPU for large dataset handling.

Certified for sustained data-centre operation

NVIDIA-Certified Systems validation confirms the platform meets thermal and reliability targets for continuous data-centre deployment. The barebone format requires the integrator to supply processors, memory, storage and operating system. Workloads that demand more than eight GPUs or a different interconnect topology will need a separate architecture.

Highlights

  • NVIDIA-Certified System for data-centre reliability
  • Eight NVIDIA HGX A100 SXM4 GPUs for massive parallel compute
  • 600 GB/s GPU-to-GPU bandwidth via NVLink and NVSwitch
  • Dual AMD EPYC 7003 processors with 3D V-Cache technology
  • 32 DDR4 DIMM slots supporting eight-channel memory per CPU

Specifications

No published specifications for this item yet.

Questions about this item

How many DDR4 DIMM slots are available for memory expansion?

The system provides 32 DIMM slots across both processors, each supporting 8-channel RDIMM or LRDIMM DDR4 memory.

What GPU-to-GPU interconnect bandwidth does the HGX A100 platform deliver?

The NVIDIA HGX A100 with 8 SXM4 GPUs uses NVLink and NVSwitch technology to achieve up to 600GB/s GPU-to-GPU interconnection.

Which processor series is required to populate the dual-socket configuration?

The barebone requires AMD EPYC 7003 Series processors, including models with 3D V-Cache technology, for the dual-processor 7nm platform.

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.