NVIDIA H100 CNX PCIe

NEW

Unprecedented performance for GPU-powered, IO-intensive workloads

Experience the unprecedented performance of converged acceleration. NVIDIA H100 CNX combines the power of the NVIDIA H100 Tensor Core GPU with the advanced networking capabilities of the NVIDIA ConnectX®-7 smart network interface card (SmartNIC) to accelerate GPU-powered, input/output (IO)-intensive workloads, such as distributed AI training in the enterprise data centre and 5G processing at the edge.

 

XENON NVIDIA DGX H100 PCIE
Specifications
NVIDIA H100 CNX PCIe
Specifications

GPU Memory

NVIDIA H100 CNX PCIe

80GB HBM2e

Specifications

Memory Bandwidth

NVIDIA H100 CNX PCIe

> 2.0TB/s

Specifications

MIG instances

NVIDIA H100 CNX PCIe

  • 7 instances @ 10GB each
  • 3 instances @ 20GB each
  • 2 instances @ 40GB each
Specifications

Interconnect

NVIDIA H100 CNX PCIe

PCIe Gen5 128GB/s

Specifications

NVLINK Bridge

NVIDIA H100 CNX PCIe

Two-way

Specifications

Networking

NVIDIA H100 CNX PCIe

1x 400Gb/s, 2x 200Gb/s ports, Ethernet or InfiniBand

Specifications

Form Factor

NVIDIA H100 CNX PCIe

Dual-slot full-height, full length (FHFL)

Specifications

Max Power

NVIDIA H100 CNX PCIe

350W

Quick Quote Request

  • This field is for validation purposes and should be left unchanged.

Get a Quote Talk to a Solutions Architect