Accelerate HPC and AI Workloads Today
Accelerate HPC and AI Workloads Today
The NVIDIA H100 Tensor Core GPU enables an order-of-magnitude leap for large-scale AI and HPC. With NVIDIA AI Enterprise for streamlined AI development and deployment, NVIDIA NVLINK Switch System direct communication between up to 256 GPUs, H100 accelerates everything from exascale scale workloads with a dedicated Transformer Engine for trillion parameter language models, down to right-sized Multi-Instance GPU (MIG) partitions.
Test drive Supermicro systems with H100 GPUs to experience an end-to-end AI and HPC data center platform.
Test drive Supermicro systems with H100 GPUs to experience an end-to-end AI and HPC data center platform.
Quick time-to-value
Turnkey solution with NVIDIA AI Enterprise and NVIDIA H100 integrated NVIDIA-Certified Systems for an AI-ready platform.
Boost performance and efficiency
Fastest AI training on large models with H100’s Transformer Engine and highest efficiency and security with 2nd gen MIG and confidential computing.
Scale with ease
Accelerate a wide range of use cases with the flexibility to service the entire AI pipeline on one infrastructure.
Enterprise ready
Streamline enterprise AI with guaranteed support, certified to run on broadly adopted enterprise platforms in the data center.
SPECIFICATION OF THE TESTING ENVIRONMENT
In cooperation with NVIDIA and Supermicro we offer, as an Elite Partner, the NVIDIA H100 PCIe GPU Test Drive Programme. Experience free of charge and without obligation in our test centre what the H100 can do in a PCIe 5.0 environment with fast storage. Interested in a test? Please fill out our request form.
Model name CPU TDP RAM GPUs NVMe Network IPMI Form factor Fields of application Power connection |
Data |
|---|
A+ Server AS-4125GS-TNRT 2x 9654 Dual Socket SP5 (AMD EPYC 9004 Series) Total 192 cores | 384 threads 350W per CPU 768GB RAM DDR5-4800 MHz 2x NVIDIA H100 Tensor Core GPU 80GB 1x 1.9TB M.2 Gen 4.0 NVMe SSD 2x 3.84TB U.2 Gen 4.0 NVMe SSD 2x 10GbE RJ45 IPMI 2.0 over LAN and KVM-over-LAN support 4 height units AI / Deep Learning | High Performance Computing 4x 2200W redundant power supplies |
.jpg)
NVIDIA H100 PCIe
Unprecedented performance, scalability, and security for mainstream servers
The NVIDIA H100 Tensor Core GPU enables a giant leap for large-scale AI performance, scalability and security for any data centre and includes the NVIDIA AI Enterprise software suite to optimise AI development and deployment. With the NVIDIA NVLINK Switch System for direct communication between up to 256 GPUs, the H100 accelerates exascale workloads with a dedicated Transformer Engine for language models with trillions of parameters. For small tasks, the H100 can be partitioned to right-side multi-instance GPU (MIG) partitions. With Hopper Confidential Computing, this scalable computing power can secure sensitive applications in a shared data centre infrastructure. NVIDIA AI Enterprise Software Suite shortens development time development and simplifies deployment of AI workloads, making the H100 the most powerful end-to-end AI and HPC data centre platform.
HIGHEST AI AND HPC MAINSTREAM PERFORMANCE
3.2PF FP8 (5X) | 1.6PF FP16 (2.5X) | 800TF TF32 (2.5X) | 48TF FP64 (2.5X)
6X faster Dynamic Programming with DPX Instructions
2TB/s , 80GB HBM2e memory
HIGHEST COMPUTE ENERGY EFFICIENCY
Configurable TDP - 150W to 350W
2 Slot FHFL mainstream form factor
HIGHEST UTILIZATION EFFICIENCY AND SECURITY
7 Fully isolated & secured instances, guaranteed QoS
2 nd Gen MIG | Confidential Computing
HIGHEST PERFORMING SERVER CONNECTIVITY
128GB/s PCI Gen5
600 GB/s GPU-2-GPU connectivity (5X PCIe Gen5)
up to 2 GPUs with NVLink Bridge
3.2PF FP8 (5X) | 1.6PF FP16 (2.5X) | 800TF TF32 (2.5X) | 48TF FP64 (2.5X)
6X faster Dynamic Programming with DPX Instructions
2TB/s , 80GB HBM2e memory
HIGHEST COMPUTE ENERGY EFFICIENCY
Configurable TDP - 150W to 350W
2 Slot FHFL mainstream form factor
HIGHEST UTILIZATION EFFICIENCY AND SECURITY
7 Fully isolated & secured instances, guaranteed QoS
2 nd Gen MIG | Confidential Computing
HIGHEST PERFORMING SERVER CONNECTIVITY
128GB/s PCI Gen5
600 GB/s GPU-2-GPU connectivity (5X PCIe Gen5)
up to 2 GPUs with NVLink Bridge
Try Before You Buy
Try Before You Buy
Test drive Supermicro systems with H100 GPUs to experience unprecedented performance, scalability and security for every data center.