Program Overview#
The NVIDIA-Certified Systems™ program has assembled the industry’s most complete set of accelerated workload performance tests to help its partners deliver the highest-performing systems. NVIDIA-Certified Systems are tested with the most powerful enterprise NVIDIA GPUs and networking and are evaluated by NVIDIA engineers for performance, functionality, scalability, and security. NVIDIA-Certified Systems have been proven to deliver predictable performance. They enable enterprises to quickly deploy optimized platforms for AI, data analytics, HPC, high-density virtual desktop infrastructure (VDI), and other accelerated workloads. These workloads run in the data center, at the edge, and on the desktop.
NVIDIA has expanded the NVIDIA-Certified Systems program beyond servers designed for the data center to include GPU-powered workstations, high-density VDI systems, and edge devices. NVIDIA-Certified Systems for the data center are tested both as single nodes and in a two-node configuration. Workstations, high-density VDI, and edge systems in the NVIDIA-Certified Systems program are evaluated on their standalone performance with the NVIDIA GPUs within a single system.
Guidelines for configuring NVIDIA-Certified Systems to achieve the best results for running a range of accelerated computing workloads can be found in the NVIDIA-Certified Systems Configuration Guide.
The NVIDIA NGC™ Catalog is the hub for GPU-accelerated and network-optimized software for AI and other compute-intensive workloads. It simplifies deployments and shortens time to solution with curated containers, pretrained models, resources, SDKs, and Helm charts. NGC software runs on a wide variety of NVIDIA GPU-accelerated platforms, including NVIDIA DGX™ Systems, on-premises servers from NVIDIA partners, and leading cloud platforms. NVIDIA-Certified Systems are tested using software from the NGC Catalog, and customers can choose to purchase enterprise-grade support through NVIDIA AI Enterprise Support Services. NVIDIA AI Enterprise Support Services provides direct access to NVIDIA subject matter experts who can quickly address software issues to minimize downtime and maximize utilization and user productivity.
NVIDIA-Certified Systems for the data center are tested with NVIDIA networking but can be purchased and deployed by customers with their choice of networking. NVIDIA partners resell NVIDIA AI Enterprise Support Services. These services are available for NVIDIA-Certified Systems configured with any network adapters and deployed in any data center.
Visit the following pages for additional information:
NVIDIA-Certified Systems Testing#
NVIDIA-Certified Systems have successfully completed a rigorous suite of functional and performance tests. NVIDIA-Certified Systems intended for cluster deployment are tested both as single nodes and in a two-node configuration. Inference Reference Configurations, workstations, high-density VDI systems, and edge devices in the NVIDIA-Certified Systems program are evaluated for standalone performance using the NVIDIA GPUs within a single system.
NVIDIA-Certified Systems configured as Inference Reference Configurations, workstations, high-density VDI systems, and edge devices run workloads such as:
Single and multi-GPU deep learning training using PyTorch
High-volume, low-latency inference using NVIDIA TensorRT™ and NVIDIA Triton™
GPU-accelerated data analytics and machine learning using NVIDIA RAPIDS™
Application development using the NVIDIA CUDA® Toolkit and the NVIDIA HPC SDK
Inference benchmarking using vLLM
NVIDIA-Certified Systems intended for cluster deployment run the workloads listed above in addition to the following:
Single-GPU and multi-GPU large language model (LLM) pretraining benchmarks based on PyTorch
Multi-node LLM pretraining
Network bandwidth testing to evaluate remote direct memory access (RDMA) performance
NVIDIA-Certified Systems Supported Software#
NVIDIA-Certified Systems Testing Software Environment#
NVIDIA-Certified Systems are tested in a standardized software environment that provides the highest levels of stability and performance. The current software test environment for NVIDIA-Certified Systems uses production versions of the following:
Ubuntu 24.04
NVIDIA drivers
DOCA SDK network software
NVIDIA Cloud Native Stack with NVIDIA GPU Operator and Kubernetes
NVIDIA also supports its AI software running on DGX OS, Red Hat Enterprise Linux, and CentOS. More details about OS support can be found at NVIDIA Optimized Frameworks Support Matrix.
NVIDIA AI Enterprise Compatible Systems#
One of the greatest values of NVIDIA-Certified Systems is that they provide a tested platform for deploying high-performance accelerated software, including NVIDIA’s own generative AI suite, NVIDIA AI Enterprise. NVIDIA AI Enterprise is an end-to-end, cloud-native software platform that accelerates data science pipelines and streamlines the development and deployment of production-grade copilots and other generative AI applications.
NVIDIA AI Enterprise is supported when used in bare-metal deployments, virtualized deployments, and in the cloud. The supported bare-metal, virtualized, and cloud software platforms are listed in the NVIDIA AI Enterprise Infrastructure Support Matrix for supported software stacks.
For on-premises deployments (bare metal or virtualized), NVIDIA AI Enterprise is supported on NVIDIA-Certified Data Center and Edge Systems. The NVIDIA-Certified Data Center Servers certified for NVIDIA AI Enterprise are listed in the List of NVIDIA-Certified Systems - Data Center Servers section. The NVIDIA-Certified Edge Systems certified for NVIDIA AI Enterprise are listed in the List of NVIDIA-Certified Systems - Edge Systems section.
The NVIDIA AI Enterprise release documentation provides detailed information on the deployment architectures certified for use with each NVIDIA AI Enterprise release. These architectures include bare-metal operating systems, virtualized software environments, and public clouds.
Refer to the documentation provided by NVIDIA’s operating system, virtualization, and cloud partners to learn which NVIDIA-Certified Systems, NVIDIA GPUs and networking devices, and third-party software they support.