NVIDIA – The best performance for AI.

Extensive computing power is required in order to be able to take advantage of AI, and one system that can meet this challenge is the NVIDIA® DGX™ based on the high performance NVIDIA GPU platform. Best of all, NVIDIA DGX offers companies GPU-optimised software and simplified management all in one compact system.

Benefits of DGX Station and DGX Server:

Upload Icon

Integrated hardware and software.

Performance Icon

Based on NVIDIA™ Tensor Core technology to accelerate AI and HPC.

Bechtle Icon

Deep learning training, inference and acceleration of analyses in one single system.

Rocket Icon

Unmatched performance for faster iterations and innovations.

DGX in the data centre...

NVIDIA DGX™ A100 is the universal system for all AI workloads, delivering unprecedented computing density, performance and flexibility in the world’s first 5 petaFLOPS AI system. NVIDIA DGX A100 features the world’s most advanced accelerator, the NVIDIA A100 Tensor Core GPU, which enables organisations to consolidate training, inference and analytics into a unified, easy-to-deploy AI infrastructure with direct access to NVIDIA AI experts.

PDF DOWNLOAD

...or in the office.

NVIDIA® DGX Station A100 is the compact workstation-sized system delivering the latest technologies, rapid deployment and incredible local computing power. To help you get the job done, a whisper-quiet cooling system has been integrated and a host of tools are available to accelerate development.

PDF DOWNLOAD

Which GPUs offer the best performance?

A100.

There is one processor that is particularly suitable for AI and that’s the NVIDIA® A100 delivers unprecedented acceleration at any scale in deep learning training and inference, data analytics and HPC.

With third-generation Tensor compute engines, the new Ampere architecture enables a 20x performance increase in the AI environment, while the new Multi-Instance GPU (MIG) feature enables and optimises workload distribution across up to seven instances, allowing data science teams to work as efficiently as possible.

T4.

The compact form factor NVIDIA® T4 is a real all-rounder. This accelerator delivers very high density and flexibility and is optimised for deep learning inference meaning it perfectly complements deep learning training systems.

In terms of graphics, the T4 delivers a native user experience for virtual office desktops as well as for virtual workstations in the mid-range CAD segment.

A40.

The NVIDIA® A40 is the best choice for virtual workstations. This accelerator delivers enough power for highly demanding applications such as CAD, simulations, VR and even enables real-time rendering for a photo-realistic result.

Choose the right NVIDIA Data Center GPU

Workload	Description	NVIDIA A100 SXM4	NVIDIA A100 PCIe	NVIDIA T4	NVIDIA A40
Workload		World's Most Powerful Data Center GPU		Versatile Data Center GPU for Mainstream Computing	World's Most Powerful Data Center GPU for Visual Computing
Deep Learning Training	For the absolute fastest model training time	8-16 GPUs for new installs	4-8 GPUs
Deep Learning Inference	For batch and real-time inference	1 GPU w/ MIG	1 GPU w/ MIG	1 GPU
HPC / AI	For scientific computing centers and higher ed and research institutions	4 GPUs with MIG for supercomputing centers	1-4 GPUs with MIG for higher ed and research
Render Farms	For batch and real-time rendering				4-8 GPUs
Graphics	For the best graphics performance on professional virtual workstations			2–8 GPUs for mid-range virtual WS or mainstream graphics	4-8 GPUs for highest performing graphics and simulation
Enterprise Acceleration	Mixed Workloads – Graphics, ML, DL, analytics, training, inference	1-4 with MIG for compute intensive multi-GPU workloads	1-4 GPUs with MIG for compute intensive single GPU workloads	4-8 GPUs for balanced workloads	2-4 GPUs for graphics intensive and compute workloads
Edge Acceleration	Edge solutions with differing use cases and location		1-2 GPU with MIG	1-8 GPUs for inference and video workloads	2-4 GPUs for graphics intensive & AR/VR workloads

Workload	Description	NVIDIA A100 SXM4	NVIDIA A100 PCIe
Workload		World's Most Powerful Data Center GPU
Deep Learning Training	For the absolute fastest model training time	8-16 GPUs for new installs	4-8 GPUs
Deep Learning Inference	For batch and real-time inference	1 GPU w/ MIG	1 GPU w/ MIG
HPC / AI	For scientific computing centers and higher ed and research institutions	4 GPUs with MIG for supercomputing centers	1-4 GPUs with MIG for higher ed and research
Render Farms	For batch and real-time rendering
Graphics	For the best graphics performance on professional virtual workstations
Enterprise Acceleration	Mixed Workloads – Graphics, ML, DL, analytics, training, inference	1-4 with MIG for compute intensive multi-GPU workloads	1-4 GPUs with MIG for compute intensive single GPU workloads
Edge Acceleration	Edge solutions with differing use cases and location		1-2 GPU with MIG

Workload	Description	NVIDIA T4	NVIDIA A40
Workload		Versatile Data Center GPU for Mainstream Computing	World's Most Powerful Data Center GPU for Visual Computing
Deep Learning Training	For the absolute fastest model training time
Deep Learning Inference	For batch and real-time inference	1 GPU
HPC / AI	For scientific computing centers and higher ed and research institutions
Render Farms	For batch and real-time rendering		4-8 GPUs
Graphics	For the best graphics performance on professional virtual workstations	2–8 GPUs for mid-range virtual WS or mainstream graphics	4-8 GPUs for highest performing graphics and simulation
Enterprise Acceleration	Mixed Workloads – Graphics, ML, DL, analytics, training, inference	4-8 GPUs for balanced workloads	2-4 GPUs for graphics intensive and compute workloads
Edge Acceleration	Edge solutions with differing use cases and location	1-8 GPUs for inference and video workloads	2-4 GPUs for graphics intensive & AR/VR workloads

* mandatory fields

Please read our Privacy Policy for information on how we process your data and protect your rights as da data subject.