NVIDIA-Certified Systems

Systems certified by NVIDIA for modern AI workloads from leading server partners.

AI is Transforming Enterprise Computing

Enterprises today need a cohesive computing infrastructure that provides the functionality, performance, security, and scalability to run modern AI workloads to deliver cutting-edge products and services, while increasing their operational efficiencies. NVIDIA-Certified Systems, comprising both NVIDIA HGX and EGX server platforms, enable enterprises to confidently deploy scalable hardware and software solutions that securely and optimally run their AI workloads.

Benefits

Performance Optimized

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

Highly Scalable

NVIDIA-Certified Systems scale out to multiple systems in a cluster with high-speed NVIDIA® Mellanox® networking solutions. High-bandwidth input/output (IO) technologies, such as GPUDirect® Storage (GDS) and GPUDirect RDMA (GDR), enable faster time to solution. 

Built-in Security

Transport Layer Security (TLS) and Internet Protocol Security (IPsec) in-line cryptography secure both customer data and AI models. Hardware root of trust enables security at the platform, data, and application layers with secure boot and secure firmware updates.

Enterprise-Grade Support

NVIDIA offers direct access to subject matter experts with NGC Support Services, so enterprises can minimize system downtime during infrastructure deployment and maximize user productivity during development.

Enabling the Secure, Accelerated Hybrid Cloud

By combining network and compute elements, an NVIDIA-Certified System can handle exploding volumes of data, enable complex AI models and applications, and support the full diversity of modern AI. NVIDIA-Certified Systems create the essential platform for accelerating the transformation of enterprise data centers. The program guides the development of AI-first infrastructure and delivers the necessary solutions for running optimized workloads both within the data center and at the edge.

NVIDIA-Certified Systems For All Workloads

NVIDIA Certified Systems for Workloads Like HPC, Machine Learning, Pro Viz, Training and Inference, and Infrastructure.

NVIDIA-Certified Systems

NVIDIA is working with its partners to certify HGX and EGX servers in single-node and cluster configurations against a diverse range of workloads, from deep learning training and inference to data analytics and industry-specific workloads.

Boxx logo
Dell Technologies logo
GIGABYTE  logo
H3C logo
Hewlett Packard Enterprise logo
Inspur logo
Supermicro logo

Get the Most From Your Systems

An NVIDIA-Certified System—validated for performance, functionality, scalability, and security—allows enterprises to easily deploy complete solutions for AI workloads from the NVIDIA NGC catalog, backed by NVIDIA NGC Support Services.

NVIDIA NGC Catalog

NGC Catalog

The NGC catalog is the hub for GPU-accelerated software that simplifies deployments with containers and increases productivity and shortens time-to-solution with pre-trained models, resources, SDKs, and Helm charts.

NGC Support Services

NGC Support Services

NGC Support Services provides enterprise-grade support to ensure NVIDIA-Certified Systems run optimally, maximizing system utilization and user productivity. The service gives enterprise IT direct access to NVIDIA subject matter experts to quickly address software issues and minimize system downtime.

Resource Center

Data Sheet for NVIDIA-Certified Systems

NVIDIA-Certified Systems

NVIDIA Press Release

Global Technology Leaders Adopt NVIDIA EGX Edge AI Platform to Infuse Intelligence at the Edge of Every Business

NVIDIA-Certified Systems Frequently Asked Questions

  • What is an NVIDIA-Certified System?

    An NVIDIA-Certified System conforms to NVIDIA’s best design practices and passes a set of certification tests that cover a range of use cases in a cluster environment. System certification is introduced with the NVIDIA® A100 GPU, the latest Mellanox network adapters (ConnectX-6, ConnectX-6 Dx , and BlueField-2 DPU), the Quantum family of InfiniBand switches, and the Spectrum family of Ethernet switches and cables to address the need to run AI workloads at scale. 

  • Why should I consider buying an NVIDIA-Certified System?

    AI is changing rapidly driven by the explosion of data from users, applications and devices. Additionally, AI models are increasing in size and complexity that simply cannot be addressed on single-node systems. NVIDIA-Certified Systems undergo comprehensive multi-node testing, so you can run your AI applications at scale, reducing the time to the solution.

  • What type of workloads can I run on the certified systems?

    An NVIDIA-Certified System can run modern workloads including data analytics, AI training and inference, professional visualization and more. The systems are tested on workloads that comprise of deep learning (DL) training, AI inference, data science algorithms, intelligent video analytics (IVA), and security, as well as network and storage offload on both single-node and cluster-based systems. The goal of the tests is to ensure that the systems can deliver the necessary scalability and performance when running real-life workloads.

  • Are NVIDIA-Certified Systems supported?

    Yes, every NVIDIA-Certified System qualifies for optional enterprise support that can be purchased directly from your certified OEM. Even if you don’t purchase the optional enterprise support, NVIDIA offers free support via our developer forum. 

     

    To learn more about support, please visit our support services page.

  • How is the NGC-Ready System different from an NVIDIA-Certified System?

    NGC-Ready was a validation program for single-node systems with NVIDIA V100, T4 and RTX 6000/8000 GPUs. 

     

    NVIDIA-Certified Systems is a forward looking certification program for both single and cluster-based systems to run AI workloads at scale, consisting of NVIDIA Ampere GPUs, NVIDIA Mellanox ConnectX-6, Connect X-6 Dx and BlueField-2 DPU.

Questions about NVIDIA-Certified Systems?