NVIDIA H100 NVL 94GB PCIe Accelerator for HPE

Technical Specifications

Detailed product description

NVIDIA H100 NVL 94GB PCIe Accelerator

Platform supported

• HPE Mainstream Compute Platforms

Number of accelerators per card

• 1

Memory size per board

• 94 GB HBM2e

Memory bandwidth for board

• 3.9 TB/s

Accelerator applications

• Transformational AI training, real-rime deep learning inference

Architecture features

• Built with 80 billion transistors
  using a cutting-edge TSMC 4N
  process custom tailored for
  NVIDIA’s accelerated compute needs, H100 NVL
  features major advances to accelerate AI,
  HPC, memory bandwidth, interconnect, and
  communication at data center scale.
  The Transformer Engine uses
  software and Hopper Tensor
  Core technology designed to
  accelerate training for models
  built from the world’s most important AI
  model building block, the transformer. Hopper
  Tensor Cores can apply mixed FP8 and FP16
  precisions to dramatically accelerate AI
  calculations for transformers.
  The NVLink Switch System
  enables the scaling of multi-
  GPU input/output (IO) across
  multiple servers at 900
  gigabytes per second (GB/s) bidirectional per
  GPU, over 7X the bandwidth of PCIe Gen5.
  The system delivers 9X higher bandwidth
  than InfiniBand HDR on the NVIDIA Ampere
  architecture.

System

• Compatible with HPE DL380 Gen10 Plus, HPE DL380 Gen11, HPE DL380a Gen11, HPE DL385 Gen11, HPE Edge Server e930t Gen11, HPE Scale-up Server 3200, HPE Superdome Flex 280, HPE DL380 Gen12, HPE DL380a Gen12

Product dimensions (imperial)

• 1.37 x 10.51 x 4.37 in

Weight

• 3.726 lb

Warranty

• For details on HPE Qualified Options Limited Warranty visit: 3-year parts, 0-year labor, and 0-year on-site support coverage. For more warranty information refer to https://support.hpe.com/hpesc/public/docDisplay?docId=sd00004309en_us