NVIDIA AI Enterprise 2.0 or later

Red Hat and NVIDIA have partnered to unlock the power of AI for every business by delivering an end-to-end enterprise platform optimized for AI workloads. This integrated platform delivers best-in-class AI software, the NVIDIA AI Enterprise suite, optimized and certified for Red Hat OpenShift, the industry’s leading Containers and Kubernetes platform. Running on NVIDIA-Certified Systems™, industry-leading accelerated servers, this platform accelerates the speed at which developers can build AI and high-performance data analytics, enabling organizations to scale modern workloads on the same infrastructure they have already invested in, and delivers enterprise-class manageability, security, and availability. Furthermore, with Red Hat OpenShift, enterprises have the flexibility to deploy in either bare-metal or virtualized environments with VMware vSphere.


The benefits of this joint NVIDIA and Red Hat AI-Ready Platform solution are:

Ease Of Deployment and Scaling

Enterprises can confidently deploy and scale an end-to-end AI solution certified by NVIDIA and Red Hat on NVIDIA-Certified Systems. This includes the flexibility to consistently deploy the NVIDIA AI Enterprise and data analytics software on Red Hat OpenShift, on bare metal or existing VMware vSphere across data centers and edge. Red Hat OpenShift’s integration with NVIDIA GPUs using the certified Kubernetes Operator and containerized AI software derisks deployments and enables seamless scaling.

Empower self-service access to AI tools and infrastructure

NVIDIA AI Enterprise on Red Hat OpenShift enables self-service, consistent, cloud-like experience for data scientists, ML engineers, and developers with the flexibility and portability to use the containerized AI tools and infrastructure resources. This allows them to rapidly build, scale, reproduce, and share the models before production rollout. They also have access to out of the box trusted, tried, tested AI tools to increase productivity and achieve faster time to value.

Secure delivery of intelligent applications with integrated MLOps

Extending OpenShift DevOps and GitOps automation capabilities to the entire AI lifecycle allows better collaboration between data scientists, ML engineers, software developers, and IT operations. This enables organizations to automate and simplify the iterative process of integrating models into software development processes, production rollout, monitoring, retraining, and redeployment for continued prediction accuracy.

When NVIDIA AI Enterprise is running on virtualized infrastructure, a key component is NVIDIA virtual GPU. A high-level architecture of an NVIDIA virtual GPU enabled environment is illustrated in the figure below. Here, we have GPUs in the server, and the NVIDIA AI Enterprise Host Software packaged as a vSphere Installation Bundle (VIB) is installed on the host server.


Red Hat OpenShift can run on VMware vSphere with GPUs in pass-through mode, which does not require the VIB.

This software enables multiple VMs to share a single GPU, or if there are multiple GPUs in the server, they can be aggregated so that a single VM can access multiple GPUs. Physical NVIDIA GPUs can support multiple virtual GPUs (vGPUs) and be assigned directly to guest VMs under the control of NVIDIA’s AI Enterprise Host Software running in a hypervisor. Guest VMs use the NVIDIA vGPUs in the same manner as a physical GPU that has been passed through by the hypervisor. In the VM itself, vGPU drivers are installed which support the different license levels that are available.


The NVIDIA AI Enterprise Host Software host driver is the software driver that is installed on the hypervisor host that is responsible for communicating with the NVIDIA vGPU guest driver which is installed on the guest VM.

GPU Operator enables DevOps teams to manage the lifecycle of GPUs when used with Red Hat OpenShift at a Cluster level. There is no need to manage each node individually. When the GPU Operator is used with Red Hat OpenShift, infrastructure teams can easily manage GPU and CPU nodes from a single pane of glass. The GPU Operator allows customers to run GPU accelerated applications on immutable operating systems as well. Faster node provisioning is achievable since the GPU Operator has been built in a way that it detects newly added GPU accelerated Kubernetes worker nodes. Then automatically installs all software components required to run GPU accelerated applications. The GPU Operator is a single tool to manage all Kubernetes components (GPU Device Plugin, GPU Feature Discovery, GPU Monitoring Tools, NVIDIA Runtime). It is important to note, GPU Operator installs NVIDIA AI Enterprise Guest Driver as well.


The components are as follows:

  • GPU Feature Discovery, which labels the worker node based on the GPU specs. This enables customers to more granularly select the GPU resources that their application requires.

  • The NVIDIA AI Enterprise Guest Driver

  • Kubernetes Device Plugin, which advertises the GPU to the Kubernetes scheduler

  • NVIDIA Container Toolkit - allows users to build and run GPU accelerated containers. The toolkit includes a container runtime library and utilities to automatically configure containers to leverage NVIDIA GPUs.

  • Data Center GPU Manager (DCGM) Monitoring - Allows monitoring of GPUs on Kubernetes.

The NVIDIA Network Operator leverages Kubernetes custom resources and the Operator framework to configure fast networking, RDMA, and GPUDirect.


To access the NVIDIA AI Enterprise Host Software (VIB), you must first download and install NGC Catalog CLI. After the NGC Catalog CLI is installed, you will need to launch a command window and then run commands to download software. It is recommended that the NGC CLI is installed on the same machine that may be used to interact with the OpenShift Cluster or ESXi hosts.

To install NGC Catalog CLI:

  1. Enter the NVIDIA NGC website

  2. In the top right corner, click Welcome and then select Setup from the menu.

  3. Click Downloads under Install NGC CLI from the Setup page.

  4. From the CLI Install page, click the Windows, Linux, or MacOS tab, according to the platform from which you will be running NGC Catalog CLI.

  5. Follow the instructions to install the CLI.

  6. Open a Terminal or Command Prompt

  7. Verify the installation by entering ngc--version. The output should be NGC Catalog CLI x.y.z where x.y.z indicates the version.

  8. You must configure NGC CLI for your use so that you can run the commands. Enter the following command and then include your API key when prompted:


$ ngc config set Enter API key [no-apikey]. Choices: [<VALID_APIKEY>, 'no-apikey']: (COPY/PASTE API KEY) Enter CLI output format type [ascii]. Choices: [ascii, csv, json]: ascii Enter org [no-org]. Choices: ['no-org']: nvlp-aienterprise Enter team [no-team]. Choices: ['no-team']: no-team Enter ace [no-ace]. Choices: ['no-ace']: no-ace

After the NGC Catalog CLI is installed, you will need to launch a command window and run the following commands to download the NVIDIA AI Enterprise Host Software (VIB).


ngc registry resource download-version "nvaie/vgpu_host_driver_2_0:510.47"

Choose the correct vib for your version of ESXi

For vSphere 7.0 U2 or later:



For vSphere 6.7:



Previous Overview
Next Prerequisites
© Copyright 2024, NVIDIA. Last updated on Apr 2, 2024.