The NVIDIA DGX H100 is the foundational building block for enterprise AI, engineered to handle the most complex AI workloads including generative AI and large language models. This liquid-cooling ready version is optimized for high-density data center environments requiring advanced thermal management.
* 8x NVIDIA H100 Tensor Core GPUs delivering 32 petaFLOPS of FP8 performance.
* 640GB total GPU memory with 3.35TB/s of memory bandwidth.
* Fourth-generation NVIDIA NVLink providing 900GB/s of GPU-to-GPU communication.
* Dual Intel Xeon Platinum 8480C processors with 112 total cores.
* 2TB of system memory to support massive data processing.
* 10x NVIDIA ConnectX-7 400Gb/s network interfaces for high-speed scalability.
* 30TB of NVMe SSD storage for high-throughput data access.
* Integrated NVIDIA AI Enterprise software suite for end-to-end AI development.
* Liquid-cooling ready chassis for superior heat dissipation in dense racks.
* Standard 19-inch rackmount form factor at 8U height.
NVIDIA
NVIDIA - DGX H100 Liquid Cooling Ready - AI Infrastructure Server
For a quote, contact us at info@tropical.com
- SKU:
- DGX H100 Liquid Cooling Ready
- Weight:
- 287.60 LBS
- Width:
- 19.00 (in)
- Height:
- 14.00 (in)
- Depth:
- 35.30 (in)
Related Products
A relatable product
NVIDIA
NVIDIA - DGX H100 (8x H100 SXM5) - AI Infrastructure System
The NVIDIA DGX H100 is the foundational building block for enterprise AI, delivering the performance required for the most demanding generative AI and natural language processing tasks. This fully...
NVIDIA - DGX SuperPOD (H100) - Scalable AI Supercomputing Infrastructure
NVIDIA DGX SuperPOD with H100 systems provides leadership-class AI infrastructure designed for the most challenging training and inference workloads. It is a full-stack data center platform that...
NVIDIA
NVIDIA - DGX H100 NVSwitch Platform - AI Enterprise Infrastructure
The NVIDIA DGX platform is the gold standard for enterprise AI factory infrastructure, designed to accelerate large-scale AI workloads and generative AI development. It provides a fully integrated...
NVIDIA - DGX BasePOD (H100) - Enterprise AI Infrastructure Reference Architecture
NVIDIA DGX BasePOD is a proven reference architecture designed to scale AI infrastructure for the enterprise, providing a foundation for building AI Centers of Excellence. It combines...
NVIDIA
NVIDIA - DGX H200 AI Server Platform - AI Supercomputing Infrastructure
The NVIDIA DGX H200 is the world’s premier AI supercomputing platform, designed to handle the most complex generative AI and large language model (LLM) workloads. It integrates NVIDIA H200 Tensor...