Get the best out of your SystEMS

Enterprises need a cohesive computing infrastructure that provides the functionality, performance, security, and scalability to run modern workloads and deliver innovative products and services while increasing operational efficiency.

an NVIDIA-certified system consisting of both NVIDIA HGX and EGX server platforms - validated for performance, functionality, scalability, and security - enables enterprises to easily implement complete solutions to support AI from the NVIDIA NGC™ catalog.

NVIDIA DGX systems

The world's first supercomputers developed for AI
Inspired by the needs of deep learning and analytics, NVIDIA® DGX™ solutions consist of a powerful, fully integrated combination of innovative GPU-optimized software, superior performance, and simplified management.
The most comprehensive AI platform in the world


The latest iteration of NVIDIA DGX systems, providing a highly systemized and scalable platform to solve the biggest challenges with AI.
2023 Q1 availableMORE

DGX H100 640GB 2TB

Expand the boundaries of business innovation and optimization with NVIDIA DGX™ H100.

AI training, inference, and analytics


A range of server solutions to help you tackle the most complex AI challenges.

DGX A100 640GB 2TB

NVIDIA DGX™ A100 is the universal system
for all AI workloads


The most powerful end-to-end platform for AI supercomputing.
Massive data sets, huge models in Deep Learning, and complex simulations require multiple GPUs with extremely fast interconnects and a fully accelerated software stack. The NVIDIA HGX™ AI supercomputing platform combines the full power of NVIDIA GPUs, NVIDIA® NVLink®, NVIDIA InfiniBand networking, and a fully optimized NVIDIA AI and HPC software stack from the NVIDIA NGC™ catalog for the highest application performance. With end-to-end performance and flexibility, NVIDIA HGX enables researchers and scientists to combine simulation, data analytics, and AI to drive scientific progress.
the NVIDIA A100 Tensor Core GPU, still incredibly powerful and successful, passes the baton to the H100, which is also compute-oriented and excels at AI training, HPC, and data processing. The H100 Tensor Core GPU adds PCIe Gen5 protocol and doubles the throughput of the NVIDIA A100 Tensor Core GPU. New NVIDIA technologies include new generations of Tensor Cores, NVLink®, NVSwitch™ and multi-instance GPU. In addition, NVIDIA Confidential Computing secures sensitive applications on shared data center infrastructure. Like the previous NVIDIA Ampere generation, the NVIDIA H100 will also offer PCIe GPU and an SXM form factor, both of which come with an extensive software ecosystem from the NVIDIA NGC catalog. The new NVIDIA H100 varient is thus to be seen as a complement to the existing NVIDIA HGX range.


For the most demanding workloads, Supermicro builds the most powerful and fastest systems based on NVIDIA A100™ Tensor Core GPUs. Supermicro supports a range of customer requirements with optimized systems for the new HGX™ A100 8-GPU and HGX™ A100 4-GPU platforms. With the latest version of NVIDIA® NVLink™ and NVIDIA NVSwitch™ technologies, these systems can deliver up to 5 PetaFLOPS of performance in a single 4U system.

Entering the Exascale Era with NVIDIA H100

GIGABYTE has already begun validating and optimizing servers for the new NVIDIA H100 technology, starting with the H100 PCIe GPU. With the launch of the NVIDIA H100, GIGABYTE will have enterprise products available in 1U, 2U and 4U server options for the H100 PCIe accelerator, with many of the 2U servers capable of supporting up to eight H100 GPUs for dense levels of parallel processing. For the NVIDIA HGX H100 platform, there are plans for 2U and 4U servers for the respective 4 & 8-GPU solutions.


High-performance, secure acceleration from the data center to the edge

From the data center to the periphery, the NVIDIA EGX platform delivers the power of accelerated AI computing with an easy-to-deploy, cloud-native software stack and management service, and a vast ecosystem of partners offering products and services with and for EGX.
EGX Server Image

End-To-End Hardware

The EGX hardware portfolio ranges from NVIDIA-Certified Systems™ for running real-time speech recognition, complex business forecasting, immersive graphical experiences, and other modern data center workloads to the small, efficient NVIDIA® Jetson™ family of products for tasks such as image recognition and sensor fusion at the edge.nVIDIA converged accelerators combine the power of NVIDIA Ampere GPUs, NVIDIA Mellanox® SmartNIC, and DPU technologies to create faster, more efficient, and more secure data centers.

Enterprise AI software

The NVIDIA EGX platform includes optimized software for accelerated AI computing from the data center to the edge. With NVIDIA AI for Enterprise, enterprises access a cloudnative end-to-end suite of AI and data analytics software optimized and certified by NVIDIA to run on VMware vSphere with NVIDIA-Certified Systems. NVIDIA AI for Enterprise includes key NVIDIA technologies for rapidly deploying, managing and scaling AI workloads in the modern hybrid cloud.

NVIDIA EGX is also compatible with Red Hat OpenShiftand other leading hybrid cloud platform partners via the NVIDIA EGX stack, which includes both the NVIDIA GPU operator and the NVIDIA Network operator. This allows enterprises to standardize and automate the deployment of all required components for Kubernetes clusters.

NVIDIA Fleet Command™ can deploy and manage line-of-business applications at the edge, such as formerly complex management tasks like remotely updating system software or monitoring a site's system health.

EGX Server Image
EGX Server Image

Acceleration for enterprise applications

WITH NVIDIA EGX, enterprises can run AI together with enterprise applications. Leverage a wide range of accelerated applications, from AI to data analytics, HPC and visualization, to real-time collaborative design and simulation. The GPU Accelerated Applications Catalog lists hundreds of accelerated applications optimized for deployment on any GPU-powered on-premises, cloud, and edge system.

NVIDIA AGX systems

A new age of autonomous machines
Self-driving cars. Autonomous robots. High-performance wearable medical devices. NVIDIA AGX systems will usher in a new era of AI-based solutions, enabling next-generation applications previously considered science fiction. This is the world's first family of systems for autonomous machines that require maximum computing power for outstanding AI.


NVIDIA DRIVE™ AGX is a scalable, open platform that serves as the "AI brain" for self-driving vehicles. It enables industry-leading performance and energy-efficient computing for the development and production of feature-rich, AI-based cars, trucks, robo-taxis, and shuttle buses. DRIVE AGX systems are all based on a single architecture and take advantage of the NVIDIA DRIVE software platform to scale from congestion pilots to fully autonomous operation.


The NVIDIA® Jetson AGX Xavier™ is a unique computer that gives you unprecedented compute power to run compute-intensive applications for outstanding AI. You can now achieve the power of a GPU workstation in a compact form factor with the highest performance at 30W or less. The Jetson AGX Xavier enables the adoption of AI in integrated applications across virtually all industries. Its energy-efficient productivity and powerful SDKs make it ideal for developing robots, drones, and other autonomous machines that require maximum computing power of up to 32 TOPS for Deep Learning tasks.


The NVIDIA Clara AGX™ Developer Kit is based on an ARM-based NVIDIA Xavier-™ system on a chip and provides real-time AI and imaging for medical devices. Combining low-power embedded ARM processors with the NVIDIA Clara AGX software developer kit and NVIDIA EGX add-on software, it is easy to securely deploy and remotely control fleets of distributed medical instruments.


Enterprise Class Support

With direct access to NVIDIA subject matter experts, organizations can minimize system downtime during infrastructure deployment and maximize user productivity during deployment.

Integrated safety

Integrated TLS (Transport Layer Security) and IPsec (Internet Protocol Security) cryptography secures both customer data and AI models. Hardware RoT (Root of Trust) enables security at the platform and data and application layers with secure boot and firmware updates.

Highly scalable

Scale to multiple systems in a cluster with NVIDIA® Mellanox® high-speed network solutions High-bandwidth input/output (I/O) technologies GPUDirect® Storage (GDS) and GPUDirect RDMA (GDR) enable faster resolution time.

Optimized performance

Each NVIDIA-certified system provides a unified, accelerated AI platform that ensures NVIDIA GPU-optimized software from the NGC catalog works out of the box - simplifying development and operationalizing AI services.

Enabling a secure accelerated hybrid cloud

By combining networking and compute elements, an NVIDIA certified system can handle the growing volumes of data, enable complex AI models and applications, and provide a tool to support the full diversity of modern AI. An NVIDIA-certified system provides an indispensable platform for accelerating enterprise data center transformation. The program will lead the development of the first AI infrastructure and deliver the necessary solutions to run optimized workloads both within the data center and in the periphery.


NVIDIA, SUPERMICRO GPU SERVER & sysGen solutions get the most out of your system.
NVIDIA certified systems provide the indispensable platform for enterprise data center advancements, providing the necessary infrastructure to run a variety of accelerated workloads. The certification test suite is designed to test the performance and functionality of a configured server by running various applications representative of a variety of real-world use cases. These include Deep Learning (DL) training, AI inference, end-to-end AI frameworks including Jarvis and Clara, Data Science including Spark, Intelligent Video Analytics (IVA), High-Performance Computing (HPC) and CUDA® capabilities, and rendering. It also addresses infrastructure performance acceleration such as network and storage offload, security features, and remote management capabilities.
sysGen offers NVIDIA-certified systems that have been tested in a variety of single-node and cluster configurations and optimized for a wide range of workloads - from deep learning training and inference to data analytics and industry-specific workloads.

An NVIDIA certified system consisting of A100, NVIDIA Mellanox Smart NICs and GPUs is validated for performance, functionality, scalability and security, allowing organizations to easily implement complete solutions for the use of AI from the NVIDIA NGC catalog. Get the most out of your systems with our NVIDIA and Supermicro recommendations, with something for every project and budget!

NVIDIA certified systems - components

Deploy NVIDIA AI on a large scale
Certified servers offer a variety of options for acceleration hardware.
the following components are supported under the NVIDIA Certified Systems program.


Ampere-based GPUs
(A100, A30, A40, A10)



Mainstream Systems

Popular server models
from Supermicro

Get the best out of your systems

An NVIDIA-certified system - validated for performance, functionality, scalability, and security - enables organizations to apply complete solutions for AI workloads from the NVIDIA NGC catalog, backed by NVIDIA NGC support services.

NVIDIA AI Enterprise

NVIDIA AI Enterprise is an end-to-end, cloud-native suite of AI and data analytics software optimized and certified by NVIDIA to run on VMware vSphere with NVIDIA-certified systems.


NVIDIA Omniverse Enterprise

NVIDIA Omniverse is a revolutionary platform for virtual collaboration and real-world simulation. Globally distributed teams can accelerate their workflows with one-click interoperability between key software tools and collaborate seamlessly in a shared virtual environment.


Cloudera gets faster

Data science teams using the Cloudera Data Platform (CDP) can accelerate data engineering, analytics, and machine learning workflows using RAPIDS and NVIDIA computing across public and private clouds.


NGC Catalog

The NGC catalog is the heart for GPU-accelerated software that simplifies container deployment, increases productivity, and reduces time to solution with pre-trained models, resources, SDKs, and Helm Charts.