Powered by NVIDIA^® GPUs

Ready for AI?

Advanced Performance for the Most Extreme AI-HPC Workloads. Hyperscale, HPC and AI infrastructures - leverage NVIDIA GPUs, InfiniBand and CUDA to deliver better performance, allowing HPC and AI technologies to be run under one system architecture.

H100 & GH200 SHIPPING AS EARLY AS DECEMBER 2023!

NVIDIA Grace Hopper™ Superchips - GH200

The Trillion-Parameter Instrument of AI

The World’s First Data Center Accelerated CPU The NVIDIA Grace™ Hopper™ architecture brings together the groundbreaking performance of the NVIDIA Hopper GPU with the versatility of the NVIDIA Grace CPU in a single superchip, connected with the high-bandwidth, memory-coherent NVIDIA® NVLink® Chip-2-Chip (C2C) interconnect.

Grace Arm® Neoverse V2 CPU
NVIDIA H100 Tensor Core GPU
Up to 480GB of LPDDR5X memory with error-correction code (ECC)
Up to 96GB of HBM3
NVLink-C2C: 900GB/s of coherent memory
Up to 576GB of fast-access memory available to the GPU
Energy-efficient 1,000W module (CPU, GPU, and memory)
2U Rackmount

GET QUOTE

NVIDIA H100/H800 SXM5

Dual Socket 4th Gen Intel® Xeon® Scalable processors or AMD EPYC™ 9004 Series Processors
Up to 8TB DDR5 DRAM
PCIe 5.0 x16 CPU-to-GPU Interconnect
8 NIC for GPU direct RDMA (1:1 GPU Ratio)
Highest GPU communication using NVIDIA® NVLINK™ + NVIDIA® NVSwitch™
7U & 8U Rackmount

GET QUOTE