NVIDIA AI Enterprise Quick Start Guide

Minimal instructions for installing and configuring NVIDIA AI Enterprise.

About this Guide

NVIDIA AI Enterprise Quick Start Guide provides minimal instructions for a bare-metal, single-node deployment of NVIDIA AI Enterprise on a third-party NVIDIA-certified system and for using a Cloud License Service (CLS) instance to serve licenses.

If you need complete instructions for installing and configuring NVIDIA AI Enterprise, are using NVIDIA AI Enterprise in an NVIDIA vGPU deployment, or are using multiple nodes, refer to NVIDIA AI Enterprise User Guide.

If you are using Delegated License Service (DLS) instances to serve licenses, refer to NVIDIA License System User Guide.

Note: The instructions in this guide do not apply to NVIDIA® DGX™ systems. For information about how to use these systems, refer to NVIDIA DGX Systems.

1. Activating the Accounts for Getting NVIDIA AI Enterprise

After your order for NVIDIA AI Enterprise is processed, you will receive an order confirmation message from NVIDIA. This message contains information that you need for getting NVIDIA AI Enterprise and technical support from NVIDIA.
To get NVIDIA AI Enterprise and technical support from NVIDIA, you must have an NVIDIA Enterprise Account, which provides login access to the following NVIDIA web properties:
  • , which provides access to all enterprise software, services, and management tools included in NVIDIA AI Enterprise
  • NVIDIA Licensing Portal, which provides access to your entitlements and options for managing your NVIDIA AI Enterprise license servers
  • NVIDIA Enterprise Support Portal, which provides access to NVIDIA AI Enterprise support services
To activate the accounts for getting NVIDIA AI Enterprise, follow this sequence of instructions:
  1. Instructions for creating or linking to an NVIDIA Enterprise Account:
  2. Setting Up Your Access to the Enterprise Catalog
  3. index.html#activating-nvidia-enterprise-account

1.1. Before You Begin

Before following the procedures in this guide, ensure that the following prerequisites are met:

For information about supported hardware and software, and any known issues for this release of NVIDIA AI Enterprise, refer to NVIDIA AI Enterprise Release Notes.

1.2. Your Order Confirmation Message

After your order for NVIDIA AI Enterprise is processed, you will receive an order confirmation message to which your NVIDIA Entitlement Certificate is attached. Your NVIDIA Entitlement Certificate contains your product activation keys and provides instructions for using the certificate.

If you are a data center administrator, follow the instructions in the NVIDIA Entitlement Certificate to use the certificate. Otherwise, forward your order confirmation message, including the attached NVIDIA Entitlement Certificate, to a data center administrator in your organization.

1.3. NVIDIA Enterprise Account Requirements

To get NVIDIA AI Enterprise, you must have a suitable NVIDIA Enterprise Account for getting NVIDIA AI Enterprise and technical support from NVIDIA.

Whether or not you have a suitable NVIDIA Enterprise Account depends on whether you have previously purchased NVIDIA AI Enterprise.
  • If you have previously purchased NVIDIA AI Enterprise, you already have a suitable NVIDIA Enterprise Account.

    To use this account to get NVIDIA AI Enterprise, download the software assets that you require from the collection on . For details, refer to Accessing the NVIDIA AI Enterprise Collection.

  • If you have obtained an evaluation license but have not previously purchased NVIDIA AI Enterprise, you do not have a suitable NVIDIA Enterprise Account.
    To create a suitable NVIDIA Enterprise Account, follow the Register link in the instructions for using the certificate to create an account for your purchased licenses. You can choose to create a separate account for your purchased licenses or link your existing account for an evaluation license to the account for your purchased licenses.
  • If you have not previously purchased NVIDIA AI Enterprise, you do not have a suitable NVIDIA Enterprise Account.

    To create a suitable NVIDIA Enterprise Account, follow the Register link in the instructions for using the certificate to create your account. For details, refer to Creating your NVIDIA Enterprise Account.

1.4. Creating your NVIDIA Enterprise Account

If you do not have an NVIDIA Enterprise Account, you must create an account to be able to log in to the web properties for getting NVIDIA AI Enterprise and technical support from NVIDIA.

For details of these web properties, refer to Activating the Accounts for Getting NVIDIA AI Enterprise.

If you already have an account, skip this task and go to Installing Your NVIDIA AI Enterprise License Server and License Files.

However, if you have an account that was created for an evaluation license and you want to access licenses that you purchased, you must repeat the registration process when you receive your purchased licenses. You can choose to create a separate account for your purchased licenses or link your existing account for an evaluation license to the account for your purchased licenses.
  • To create a separate account for your purchased licenses, perform this task, specifying a different e-mail address than the address with which you created your existing account.
  • To link your existing account for an evaluation license to the account for your purchased licenses, follow the instructions in Linking an Evaluation Account to an NVIDIA Enterprise Account for Purchased Licenses, specifying the e-mail address with which you created your existing account.
Before you begin, ensure that you have your order confirmation message.
  1. In the instructions for using your NVIDIA Entitlement Certificate, follow the Register link.
  2. Fill out the form on the NVIDIA Enterprise Account Registration page and click Register.
A message confirming that an account has been created appears and the following e-mail messages are sent:
  • An e-mail message is sent to the e-mail address you provided for the NGC administrator that invites the NGC administrator to access .
  • An e-mail message is sent to the e-mail address you provided for the primary contact that instructs the primary contact to set a password for the NVIDIA Enterprise Account.

1.5. Linking an Evaluation Account to an NVIDIA Enterprise Account for Purchased Licenses

If you have an account that was created for an evaluation license, you must repeat the registration process when you receive your purchased licenses. To link your existing account for an evaluation license to the account for your purchased licenses, register for an NVIDIA Enterprise Account with the e-mail address with which you created your existing account.

If you want to create a separate account for your purchased licenses, follow the instructions in Creating your NVIDIA Enterprise Account, specifying a different e-mail address than the address with which you created your existing account.

  1. In the instructions for using the NVIDIA Entitlement Certificate for your purchased licenses, follow the Register link.
  2. Fill out the form on the NVIDIA Enterprise Account Registration page, specifying the e-mail address with which you created your existing account, and click Register.

    Screen capture showing the NVIDIA Enterprise Account Registration page.

  3. When a message stating that your e-mail address is already linked to an evaluation account is displayed, click LINK TO NEW ACCOUNT.

    Screen capture showing the option for linking an evaluation account to a new NVIDIA Enterprise Account for purchased licenses.

Log in to the NVIDIA Licensing Portal with the credentials for your existing account.

1.6. Downloading NVIDIA AI Enterprise

Before you begin, ensure that you have your order confirmation message and have created an NVIDIA Enterprise Account.
  1. Visit the NVIDIA Enterprise Application Hub by following the Login link in the instructions for using your NVIDIA Entitlement Certificate or when prompted after setting the password for your NVIDIA Enterprise Account.
  2. When prompted, provide your e-mail address and password, and click LOGIN.

    Screen capture showing the NVIDIA Enterprise Account page.

  3. On the NVIDIA APPLICATION HUB page that opens, click NVIDIA LICENSING PORTAL.

    The NVIDIA Licensing Portal dashboard page opens.



    Screen capture showing the NVIDIA Licensing Portal dashboard with no license servers.

    Note: Your entitlement might not appear on the NVIDIA Licensing Portal dashboard page until 24 business hours after you set your password during the initial registration process.
  4. In the NVIDIA Licensing Portal dashboard page opens, click the down arrow next to each entitlement listed to view details of the NVIDIA AI Enterprise that you purchased.

    Screen capture showing the NVIDIA Licensing Portal dashboard with all entitlements expanded.

  5. In the left navigation pane of the NVIDIA Licensing Portal dashboard, click SOFTWARE DOWNLOADS.
  6. On the Product Download page that opens, set the Product Family option to NVAIE and follow the Download link for NVIDIA AI Enterprise.
  7. When prompted to accept the license for the software that you are downloading, click AGREE & DOWNLOAD.
  8. When the browser asks what it should do with the file, select the option to save the file.

    After the download starts, a pop-up window opens for you to download any additional software that you might need for your NVIDIA AI Enterprise deployment.

  9. In the pop-up window, follow the links to download any additional software that you need for your NVIDIA AI Enterprise deployment.
    1. If you are using Delegated License Service (DLS) instances to serve licenses, follow the link to DLS 1.0 for your chosen hypervisor, for example, DLS 1.0 for VMware vSphere. For information about installing and configuring DLS instances, refer to NVIDIA License System User Guide.

2. Accessing the Enterprise Catalog and the NGC Private Registry

2.1. The Enterprise Catalog

The NVIDIA AI Enterprise Software Suite is distributed through the Enterprise Catalog. After you access the Enterprise Catalog, you will see the NVIDIA AI Enterprise Software Suite collection. Detailed documentation makes it easy to utilize the software, and if additional support is required, users can submit the ticket directly from the portal.

2.1.1. Setting Up Your Access to the Enterprise Catalog

  1. After your access was set up, you will receive a welcome email that invites you to continue the login process. Click on Activate Account.



  2. Click on Create Account to create a new NVIDIA account. If you already have an existing NVIDIA account linked to this email address, login here.



  3. Provide account details and accept the NVIDIA Account Terms of Use. Click on Create Account.



  4. To complete your profile, you are asked to verify your account.



  5. Go to your email inbox, open the “NVIDIA Account Created” email, and click on Verify Email Address.



  6. You are redirected to the following screen. Set your recommendation settings. Click Submit.



  7. Review and accept the NVIDIA Account Terms of Use and the NVIDIA Privacy Policy.



  8. Complete your profile by providing the information below. Click Continue.



  9. Review and Accept the NVIDIA GPU Cloud Terms of Use and Consent.



  10. Review and Accept the NVIDIA AI Enterprise Terms of Use.



  11. If asked, set your organization. The name of your organization was defined while setting up your Private Registry. Click Sign In.



  12. Welcome to the Enterprise Catalog.



2.1.2. Downloading Software from the Enterprise Catalog

2.1.2.1. Accessing the NVIDIA AI Enterprise Collection

  1. Go to https://ngc.nvidia.com/catalog/enterprise and, if prompted, log in. Click on the NVIDIA AI Enterprise Collection.



  2. Click on the Entities tab to review all the software assets part of the NVIDIA AI Enterprise stack.



  3. Click on the software asset you are interested in to learn more or download the software in the entities view.



2.1.2.2. Container Images

To pull AI and data science containers using Docker, follow these steps within the VM:
  1. Generate your API key.
  2. Access the Enterprise Catalog Container Registry.
    1. Log in to the NGC container registry.
      sudo docker login nvcr.io
    2. When prompted for your username, enter the text $oauthtoken.
      Username: $oauthtoken
    3. When prompted for your password, enter your NGC API key.
      Password: my-api-key
  3. For each AI or data science application that you are interested in, load the container.
    sudo docker pull nvcr.io/nvaie/tensorflow:21.02-tf2-py3

2.1.2.3. Helm Charts

  1. Go to the Enterprise Catalog.
  2. Click on the NVIDIA AI Enterprise Collection.
  3. Go to the Entities tab and select the Helm chart you are interested in.
  4. Here is how you download a Helm chart from the Enterprise Catalog.

2.1.2.4. Resources

  1. Go to the Enterprise Catalog.
  2. Click on the NVIDIA AI Enterprise Collection.
  3. Go to the Entities tab and select the Resource you are interested in. You can either download the Resource directly from the UI or use the displayed wget or CLI commands.

2.1.3. Adding Additional Users from Your Organization to the Enterprise Catalog (Admins Only)

As an admin, you are responsible for giving members of your organization access to the Enterprise Catalog.
  1. Make sure you are signed in.
  2. Make sure to select your company's organization from the user menu on the top right.



  3. On the left side menu, select Organization and click on Users, then click the + icon at the bottom of the screen and then click the Invite New User icon.







  4. Provide the name and email address of the user you would like to add.



  5. Provision user roles for the new user:
    1. To give the new user access to the entities in the Enterprise Catalog, provide them with the user role NVIDIA AI Enterprise Viewer.



    2. To make them an admin that can add additional users to the Enterprise Catalog, provision the user roles: NVIDIA AI Enterprise Viewer and User Admin.



    3. To give the user access to your organization’s Private Registry, see Accessing Your NGC Private Registry. Provisioning access to the Enterprise Catalog and your organization’s Private Registry can be done in one or two steps.



2.2. The NGC Private Registry

As an NVIDIA AI Enterprise user, you have exclusive access to your organization’s own NGC Private Registry, which gives authorized users within your organization privileges to store your company’s proprietary software and tools, including custom models, frameworks, and helm charts, in one location.

The complete NGC Private Registry user guide can be found here.

2.2.1. Accessing Your NGC Private Registry

  1. To access your NGC Private Registry, sign in with your NGC Account.
  2. In the top right corner, click your user account icon and select the orgname.



  3. To view artifacts in your NGC Private Registry, select Private Registry in the left-hand menu.



  4. You can access the content of the NGC Private Registry by selecting one of the entity types (Collections, Containers, Helm Charts, Models, Resources).
  5. To upload entities to your NGC Private Registry, click on Entity Creation Hub.

2.2.2. Managing Teams and Users

As an admin, you can add users to your organization’s NGC Private Registry and create teams within the NGC Private Registry.

Before adding users and teams, familiarize yourself with the following definitions of each role here.

2.2.2.1. Creating Teams

Creating teams allows users to share images within a team while keeping them invisible to other teams in the same organization. Only organization administrators can create teams.

Here is how you create a team.

2.2.2.2. Creating Users

As the organization administrator, you must create user accounts to allow others to use the NGC container registry within the organization.

Here is how you create a new user.

3. Installing Your NVIDIA AI Enterprise License Server and License Files

The NVIDIA License System is used to serve a pool of floating licenses to licensed NVIDIA software products. The NVIDIA License System is configured with licenses obtained from the NVIDIA Licensing Portal.

Note: These instructions cover only the configuration of a Cloud License Service (CLS) instance. If you need complete instructions or are using Delegated License Service (DLS) instances to serve licenses, refer to NVIDIA License System User Guide.

3.1. Introduction to NVIDIA Software Licensing

To activate licensed functionalities, a licensed client must obtain a software license when it is booted.

A client with a network connection obtains a license by leasing it from a NVIDIA License System service instance. The service instance serves the license to the client over the network from a pool of floating licenses obtained from the NVIDIA Licensing Portal. The license is returned to the service instance when the licensed client no longer requires the license.

3.2. Creating a License Server on the NVIDIA Licensing Portal

To be able to allot licenses to an NVIDIA License System instance, you must create at least one license server on the NVIDIA Licensing Portal. Creating a license server defines the set of licenses to be allotted.

  1. In the NVIDIA Licensing Portal, navigate to the organization or virtual group for which you want to create the license server.
    1. If you are not already logged in, log in to the NVIDIA Enterprise Application Hub and click NVIDIA LICENSING PORTAL to go to the NVIDIA Licensing Portal.
    2. Optional: If your assigned roles give you access to multiple virtual groups, click View settings at the top right of the page and in the My Info window that opens, select the virtual group from the Virtual Group drop-down list, and close the My Info window.
    If no license servers have been created for your organization or virtual group, the NVIDIA Licensing Portal dashboard displays a message asking if you want to create a license server.
  2. In the left navigation pane of the NVIDIA Licensing Portal dashboard, expand LICENSE SERVER and click CREATE SERVER. The Create License Server wizard is started.

    Screen capture showing the Create Server tab of the NLS navigation pane

    The Create License Server wizard opens.



    Screen capture showing the Create License Server wizard

  3. On the Create License Server page of the wizard, step through the configuration requirements to provide the details of your license server.
    1. Step 1 – Identification: In the Name field, enter your choice of name for the license server and in the Description field, enter a text description of the license server. The description is required and will be displayed on the details page for the license server that you are creating.
    2. Step 2 – Features: Select one or more available features from your entitlements to allot to this license server.
    3. Step 3 - Environment: Select Cloud (CLS) or On-Premises (DLS) to install this license server. To make the selection after the license server has been created, select the Deferred option.
    4. Step 4 – Configuration: From the Leasing mode drop-down list, select one of the following leasing modes:
      Standard Networked Licensing
      Select this mode to simplify the management of licenses on a license server that supports networked licensing. In this mode, no additional configuration of the licenses on the server is required.
      Advanced Networked Licensing
      Select this mode if you require control over the management of licenses on a license server that supports networked licensing. This mode requires additional configuration to create license pools and fulfillment conditions on the server.
      Node-Locked Licensing
      Select this mode only if the license server will serve clients that cannot obtain a license from a remote license server over a network connection. In this mode, the clients obtain a node-locked license from a file installed locally on the client system.
      CAUTION:
      This mode requires additional work to create the license file to be installed locally and to return licenses when the client is shut down. If this mode is set, the mode of the license server cannot be changed.
    5. Click REVIEW SUMMARY to review the configuration summary before creating the license server.
  4. On the Create License Server page, from the Step 4 – Configuration menu, click the CREATE SERVER option to create this license server. Alternatively, you can click CREATE SERVER on the Server Summary page.

3.3. Creating a CLS Instance on the NVIDIA Licensing Portal

When you create a CLS instance, the instance is automatically registered with the NVIDIA Licensing Portal. This task is only necessary if you are not using the default CLS instance.

  1. If you are not already logged in, log in to the NVIDIA Enterprise Application Hub and click NVIDIA LICENSING PORTAL to go to the NVIDIA Licensing Portal.
  2. In the left navigation pane of the NVIDIA Licensing Portal dashboard, click SERVICE INSTANCES.

    Screen capture showing the Service Instances tab on the NLS navigation pane

  3. On the Service Instances page, from the Actions menu, choose Create cloud (CLS) instance.

    The Create cloud (CLS) instance pop-up window opens.

  4. Provide the details of your cloud service instance.
    1. In the Name field, enter your choice of name for the service instance.
    2. In the Description field, enter a text description of the service instance. This description is required and will be displayed on the Service Instances page when the entry for service instance that you are creating is expanding.
  5. Click CREATE CLS INSTANCE.

3.4. Binding a License Server to a Service Instance

Binding a license server to a service instance ensures that licenses on the server are available only from that service instance. As a result, the licenses are available only to the licensed clients that are served by the service instance to which the license server is bound.

  1. In the NVIDIA Licensing Portal, navigate to the organization or virtual group to which the license server belongs.
    1. If you are not already logged in, log in to the NVIDIA Enterprise Application Hub and click NVIDIA LICENSING PORTAL to go to the NVIDIA Licensing Portal.
    2. Optional: If your assigned roles give you access to multiple virtual groups, click View settings at the top right of the page and in the My Info window that opens, select the virtual group from the Virtual Group drop-down list, and close the My Info window.
  2. In the left navigation pane of the NVIDIA Licensing Portal dashboard, expand LICENSE SERVERS and click LIST SERVERS.
  3. In the list of license servers on the License Servers page that opens, from the Actions menu for the license server, choose Bind.
  4. In the Bind Service Instance pop-up window that opens, select the service instance to which you want to bind the license server and click BIND. The Bind Service Instance pop-up window confirms that the license server has been bound to the service instance.

3.5. Installing a License Server on a CLS Instance

This task is necessary only if you are not using the default CLS instance.
  1. In the NVIDIA Licensing Portal, navigate to the organization or virtual group for which you want to install the license server.
    1. If you are not already logged in, log in to the NVIDIA Enterprise Application Hub and click NVIDIA LICENSING PORTAL to go to the NVIDIA Licensing Portal.
    2. Optional: If your assigned roles give you access to multiple virtual groups, click View settings at the top right of the page and in the My Info window that opens, select the virtual group from the Virtual Group drop-down list, and close the My Info window.
  2. In the left navigation pane of the NVIDIA Licensing Portal dashboard, expand LICENSE SERVER and click LIST SERVERS.
  3. In the list of license servers on the License Servers page that opens, click the name of the license server that you want to install.
  4. In the License Server Details page that opens, from the Actions menu, choose Install.
  5. In the Install License Server pop-up window that opens, click INSTALL SERVER.

3.6. Generating a Client Configuration Token for a CLS Instance

  1. Log in to the NVIDIA Enterprise Application Hub and click NVIDIA LICENSING PORTAL to go to the NVIDIA Licensing Portal.
  2. If your assigned roles give you access to multiple virtual groups, select the virtual group for which you are managing licenses from the list of virtual groups at the top right of the NVIDIA Licensing Portal dashboard.
  3. In the left navigation pane, click SERVICE INSTANCES.

    Screen capture showing the Service Instances topic in the navigation pane.

  4. On the Service Instances page that opens, from the Actions menu for the CLS instance for which you want to generate a client configuration token, choose Generate client configuration token.
  5. In the Generate Client Configuration Token pop-up window that opens, select the references that you want to include in the client configuration token.
    1. From the list of scope references, select the scope references that you want to include.

      Screen capture showing the selection of scope references and fulfillment class references for inclusion in a client configuration token

      You must select at least one scope reference.

      Each scope reference specifies the license server that will fulfil a license request.

    2. Optional: Click the Fulfillment class references tab, and from the list of fulfillment class references, select the fulfillment class references that you want to include.

      Screen capture showing the selection of scope references and fulfillment class references for inclusion in a client configuration token

      Including fulfillment class references is optional.
    3. Optional: In the Expiration section, select an expiration date for the client configuration token. If you do not select a date, the default token expiration time is 12 years.
    4. Click DOWNLOAD CLIENT CONFIGURATION TOKEN.
    A file named client_configuration_token_mm-dd-yyyy-hh-mm-ss.tok is saved to your default downloads folder.

4. Installing and Configuring NVIDIA vGPU Manager

Before installing and configuring NVIDIA vGPU Manager, ensure that a VM running a supported guest OS is configured in your chosen hypervisor.

The factory settings of some supported GPU boards are incompatible with NVIDIA AI Enterprise. Before configuring NVIDIA AI Enterprise on these GPU boards, you must configure the boards to change these settings.

4.1. Switching the Mode of a GPU that Supports Multiple Display Modes

Some GPUs support display-off and display-enabled modes but must be used in NVIDIA AI Enterprise deployments in display-off mode.

The GPUs listed in the following table support multiple display modes. As shown in the table, some GPUs are supplied from the factory in display-off mode, but other GPUs are supplied in a display-enabled mode.

GPU Mode as Supplied from the Factory
NVIDIA A40 Display-off
NVIDIA L40 Display-off
NVIDIA RTX 6000 Ada Display enabled
NVIDIA RTX A5000 Display enabled
NVIDIA RTX A5500 Display enabled
NVIDIA RTX A6000 Display enabled

A GPU that is supplied from the factory in display-off mode, such as the NVIDIA A40 GPU, might be in a display-enabled mode if its mode has previously been changed.

To change the mode of a GPU that supports multiple display modes, use the displaymodeselector tool, which you can request from the NVIDIA Display Mode Selector Tool page on the NVIDIA Developer website.

Note:

Only the following GPUs support the displaymodeselector tool:

  • NVIDIA A40
  • NVIDIA L40
  • NVIDIA RTX A5000
  • NVIDIA RTX 6000 Ada
  • NVIDIA RTX A5500
  • NVIDIA RTX A6000

Other GPUs that support NVIDIA AI Enterprise do not support the displaymodeselector tool and, unless otherwise stated, do not require display mode switching.

4.2. Installing the NVIDIA Virtual GPU Manager on VMware vSphere

The NVIDIA Virtual GPU Manager runs on the ESXi host. It is distributed as a number of software components in a ZIP archive.

The NVIDIA Virtual GPU Manager software components are as follows:
  • A software component for the NVIDIA vGPU hypervisor host driver
  • A software component for the NVIDIA GPU Management daemon
Before you begin, ensure that the following prerequisites are met:
  • The ZIP archive that contains NVIDIA AI Enterprise has been downloaded from the NVIDIA Licensing Portal.
  • The software components for the NVIDIA Virtual GPU Manager have been extracted from the downloaded ZIP archive.
  1. Copy the NVIDIA Virtual GPU Manager component files to the ESXi host.
  2. Put the ESXi host into maintenance mode.
    $ esxcli system maintenanceMode set –-enable true
  3. Install the NVIDIA vGPU hypervisor host driver and the NVIDIA GPU Management daemon from their software component files.
    1. Run the esxcli command to install the NVIDIA vGPU hypervisor host driver from its software component file.
      $ esxcli software vib install -d /vmfs/volumes/datastore/host-driver-component.zip
    2. Run the esxcli command to install the NVIDIA GPU Management daemon from its software component file.
      $ esxcli software vib install -d /vmfs/volumes/datastore/gpu-management-daemon-component.zip
    datastore
    The name of the VMFS datastore to which you copied the software components.
    host-driver-component
    The name of the file that contains the NVIDIA vGPU hypervisor host driver in the form of a software component. Ensure that you specify the file that was extracted from the downloaded ZIP archive. For example, for VMware vSphere 7.0.2, host-driver-component is NVD-VMware-x86_64-525.105.14-1OEM.702.0.0.17630552-bundle-build-number.
    gpu-management-daemon-component
    The name of the file that contains the NVIDIA GPU Management daemon in the form of a software component. Ensure that you specify the file that was extracted from the downloaded ZIP archive. For example, for VMware vSphere 7.0.2, gpu-management-daemon-component is VMW-esx-7.0.2-nvd-gpu-mgmt-daemon-1.0-0.0.0001.
  4. Exit maintenance mode.
    $ esxcli system maintenanceMode set –-enable false
  5. Reboot the ESXi host.
    $ reboot
  6. Verify that the NVIDIA GPU Management daemon has started.
    $ /etc/init.d/nvdGpuMgmtDaemon status
  7. Verify that the NVIDIA kernel driver can successfully communicate with the physical GPUs in your system by running the nvidia-smi command without any options.
    $ nvidia-smi
    If successful, the nvidia-smi command lists all the GPUs in your system.

4.3. Disabling and Enabling ECC Memory

Some GPUs that support NVIDIA AI Enterprise support error correcting code (ECC) memory with NVIDIA vGPU. ECC memory improves data integrity by detecting and handling double-bit errors. However, not all GPUs, vGPU types, and hypervisor software versions support ECC memory with NVIDIA vGPU.

On GPUs that support ECC memory with NVIDIA vGPU, ECC memory is supported with C-series and Q-series vGPUs, but not with A-series and B-series vGPUs. Although A-series and B-series vGPUs start on physical GPUs on which ECC memory is enabled, enabling ECC with vGPUs that do not support it might incur some costs.

On physical GPUs that do not have HBM2 memory, the amount of frame buffer that is usable by vGPUs is reduced. All types of vGPU are affected, not just vGPUs that support ECC memory.

The effects of enabling ECC memory on a physical GPU are as follows:

  • ECC memory is exposed as a feature on all supported vGPUs on the physical GPU.
  • In VMs that support ECC memory, ECC memory is enabled, with the option to disable ECC in the VM.
  • ECC memory can be enabled or disabled for individual VMs. Enabling or disabling ECC memory in a VM does not affect the amount of frame buffer that is usable by vGPUs.

GPUs based on the Pascal GPU architecture and later GPU architectures support ECC memory with NVIDIA vGPU. To determine whether ECC memory is enabled for a GPU, run nvidia-smi -q for the GPU.

Tesla M60 and M6 GPUs support ECC memory when used without GPU virtualization, but NVIDIA vGPU does not support ECC memory with these GPUs. In graphics mode, these GPUs are supplied with ECC memory disabled by default.

Some hypervisor software versions do not support ECC memory with NVIDIA vGPU.

If you are using a hypervisor software version or GPU that does not support ECC memory with NVIDIA vGPU and ECC memory is enabled, NVIDIA vGPU fails to start. In this situation, you must ensure that ECC memory is disabled on all GPUs if you are using NVIDIA vGPU.

4.3.1. Disabling ECC Memory

If ECC memory is unsuitable for your workloads but is enabled on your GPUs, disable it. You must also ensure that ECC memory is disabled on all GPUs if you are using NVIDIA vGPU with a hypervisor software version or a GPU that does not support ECC memory with NVIDIA vGPU. If your hypervisor software version or GPU does not support ECC memory and ECC memory is enabled, NVIDIA vGPU fails to start.

Where to perform this task depends on whether you are changing ECC memory settings for a physical GPU or a vGPU.

  • For a physical GPU, perform this task from the hypervisor host.
  • For a vGPU, perform this task from the VM to which the vGPU is assigned.
    Note: ECC memory must be enabled on the physical GPU on which the vGPUs reside.
Before you begin, ensure that NVIDIA Virtual GPU Manager is installed on your hypervisor. If you are changing ECC memory settings for a vGPU, also ensure that the NVIDIA AI Enterprise graphics driver is installed in the VM to which the vGPU is assigned.
  1. Use nvidia-smi to list the status of all physical GPUs or vGPUs, and check for ECC noted as enabled.
    # nvidia-smi -q
    
    ==============NVSMI LOG==============
    
    Timestamp                           : Mon Apr 17 18:36:45 2023
    Driver Version                      : 525.105.14
    
    Attached GPUs                       : 1
    GPU 0000:02:00.0
    
    [...]
    
        Ecc Mode
            Current                     : Enabled
            Pending                     : Enabled
    
    [...]
  2. Change the ECC status to off for each GPU for which ECC is enabled.
    • If you want to change the ECC status to off for all GPUs on your host machine or vGPUs assigned to the VM, run this command:
      # nvidia-smi -e 0
    • If you want to change the ECC status to off for a specific GPU or vGPU, run this command:
      # nvidia-smi -i id -e 0

      id is the index of the GPU or vGPU as reported by nvidia-smi.

      This example disables ECC for the GPU with index 0000:02:00.0.

      # nvidia-smi -i 0000:02:00.0 -e 0
  3. Reboot the host or restart the VM.
  4. Confirm that ECC is now disabled for the GPU or vGPU.
    # nvidia—smi —q
    
    ==============NVSMI LOG==============
    
    Timestamp                           : Mon Apr 17 18:37:53 2023
    Driver Version                      : 525.105.14
    
    Attached GPUs                       : 1
    GPU 0000:02:00.0
    [...]
    
        Ecc Mode
            Current                     : Disabled
            Pending                     : Disabled
    
    [...]

4.3.2. Enabling ECC Memory

If ECC memory is suitable for your workloads and is supported by your hypervisor software and GPUs, but is disabled on your GPUs or vGPUs, enable it.

Where to perform this task depends on whether you are changing ECC memory settings for a physical GPU or a vGPU.

  • For a physical GPU, perform this task from the hypervisor host.
  • For a vGPU, perform this task from the VM to which the vGPU is assigned.
    Note: ECC memory must be enabled on the physical GPU on which the vGPUs reside.
Before you begin, ensure that NVIDIA Virtual GPU Manager is installed on your hypervisor. If you are changing ECC memory settings for a vGPU, also ensure that the NVIDIA AI Enterprise graphics driver is installed in the VM to which the vGPU is assigned.
  1. Use nvidia-smi to list the status of all physical GPUs or vGPUs, and check for ECC noted as disabled.
    # nvidia-smi -q
    
    ==============NVSMI LOG==============
    
    Timestamp                           : Mon Apr 17 18:36:45 2023
    Driver Version                      : 525.105.14
    
    Attached GPUs                       : 1
    GPU 0000:02:00.0
    
    [...]
    
        Ecc Mode
            Current                     : Disabled
            Pending                     : Disabled
    
    [...]
  2. Change the ECC status to on for each GPU or vGPU for which ECC is enabled.
    • If you want to change the ECC status to on for all GPUs on your host machine or vGPUs assigned to the VM, run this command:
      # nvidia-smi -e 1
    • If you want to change the ECC status to on for a specific GPU or vGPU, run this command:
      # nvidia-smi -i id -e 1

      id is the index of the GPU or vGPU as reported by nvidia-smi.

      This example enables ECC for the GPU with index 0000:02:00.0.

      # nvidia-smi -i 0000:02:00.0 -e 1
  3. Reboot the host or restart the VM.
  4. Confirm that ECC is now enabled for the GPU or vGPU.
    # nvidia—smi —q
    
    ==============NVSMI LOG==============
    
    Timestamp                           : Mon Apr 17 18:37:53 2023
    Driver Version                      : 525.105.14
    
    Attached GPUs                       : 1
    GPU 0000:02:00.0
    [...]
    
        Ecc Mode
            Current                     : Enabled
            Pending                     : Enabled
    
    [...]

4.4. Changing the Default Graphics Type in VMware vSphere

The vGPU Manager VIB for VMware vSphere provides vSGA and vGPU functionality in a single VIB. After this VIB is installed, the default graphics type is Shared, which provides vSGA functionality. To enable vGPU support for VMs in VMware vSphere, you must change the default graphics type to Shared Direct.

If you do not change the default graphics type, VMs to which a vGPU is assigned fail to start and the following error message is displayed:

The amount of graphics resource available in the parent resource pool is insufficient for the operation.
Note: Change the default graphics type before configuring vGPU. Output from the VM console in the VMware vSphere Web Client is not available for VMs that are running vGPU.

Before changing the default graphics type, ensure that the ESXi host is running and that all VMs on the host are powered off.

  1. Log in to vCenter Server by using the vSphere Web Client.
  2. In the navigation tree, select your ESXi host and click the Configure tab.
  3. From the menu, choose Graphics and then click the Host Graphics tab.
  4. On the Host Graphics tab, click Edit.
  5. In the Edit Host Graphics Settings dialog box that opens, select Shared Direct and click OK.

    After you click OK, the default graphics type changes to Shared Direct.

  6. Click the Graphics Devices tab to verify the configured type of each physical GPU on which you want to configure vGPU. The configured type of each physical GPU must be Shared Direct. For any physical GPU for which the configured type is Shared, change the configured type as follows:
    1. On the Graphics Devices tab, select the physical GPU and click the Edit icon.
    2. In the Edit Graphics Device Settings dialog box that opens, select Shared Direct and click OK.
  7. Restart the ESXi host or stop and restart nv-hostengine on the ESXi host.

    To stop and restart nv-hostengine, perform these steps:

    1. Stop nv-hostengine.
      [root@esxi:~] nv-hostengine -t
    2. Wait for 1 second to allow nv-hostengine to stop.
    3. Start nv-hostengine.
      [root@esxi:~] nv-hostengine -d
  8. In the Graphics Devices tab of the VMware vCenter Web UI, confirm that the active type and the configured type of each physical GPU are Shared Direct.

4.5. Configuring a vSphere VM with NVIDIA vGPU

CAUTION:
Output from the VM console in the VMware vSphere Web Client is not available for VMs that are running vGPU. Make sure that you have installed an alternate means of accessing the VM (such as VMware Horizon or a VNC server) before you configure vGPU.

VM console in vSphere Web Client will become active again once the vGPU parameters are removed from the VM’s configuration.

How to configure a vSphere VM with a vGPU depends on your VMware vSphere version as explained in the following topics:

After you have configured a vSphere VM with a vGPU, start the VM. VM console in vSphere Web Client is not supported in this vGPU release. Therefore, use VMware Horizon or VNC to access the VM’s desktop.

4.5.1. Configuring a vSphere 8 VM with NVIDIA vGPU

  1. Open the vCenter Web UI.
  2. In the vCenter Web UI, right-click the VM and choose Edit Settings.
  3. In the Edit Settings window that opens, configure the vGPUs that you want to add to the VM. Add each vGPU that you want to add to the VM as follows:
    1. From the ADD NEW DEVICE menu, choose PCI Device.
    2. In the Device Selection window that opens, select the type of vGPU you want to configure and click SELECT.
  4. Back in the Edit Settings window, click OK.

4.5.2. Configuring a vSphere 7 VM with NVIDIA vGPU

If you are adding multiple vGPUs to a single VM, perform this task for each vGPU that you want to add to the VM.
  1. Open the vCenter Web UI.
  2. In the vCenter Web UI, right-click the VM and choose Edit Settings.
  3. Click the Virtual Hardware tab.
  4. In the New device list, select Shared PCI Device and click Add. The PCI device field should be auto-populated with NVIDIA GRID vGPU.
  5. From the GPU Profile drop-down menu, choose the type of vGPU you want to configure and click OK.
  6. Ensure that VMs running vGPU have all their memory reserved:
    1. Select Edit virtual machine settings from the vCenter Web UI.
    2. Expand the Memory section and click Reserve all guest memory (All locked).

5. Installing and Licensing NVIDIA AI Enterprise Components Required in a Guest VM

5.1. Installing the NVIDIA AI Enterprise Graphics Driver on Ubuntu from a Debian Package

The NVIDIA AI Enterprise graphics driver for Ubuntu is distributed as a Debian package file.

This task requires sudo privileges.
  1. Copy the NVIDIA AI Enterprise Linux driver package, for example nvidia-linux-grid-525_525.105.17_amd64.deb, to the guest VM where you are installing the driver.
  2. Log in to the guest VM as a user with sudo privileges.
  3. Open a command shell and change to the directory that contains the NVIDIA AI Enterprise Linux driver package.
  4. From the command shell, run the command to install the package.
    $ sudo apt-get install ./nvidia-linux-grid-525_525.105.17_amd64.deb
  5. Verify that the NVIDIA driver is operational.
    1. Reboot the system and log in.
    2. After the system has rebooted, confirm that you can see your NVIDIA vGPU device in the output from the nvidia-smi command.
      $ nvidia-smi

5.2. Configuring a Licensed Client

A client with a network connection obtains a license by leasing it from a NVIDIA License System service instance. The service instance serves the license to the client over the network from a pool of floating licenses obtained from the NVIDIA Licensing Portal. The license is returned to the service instance when the licensed client no longer requires the license.

The graphics driver creates a default location in which to store the client configuration token on the client.

The process for configuring a licensed client is the same for CLS and DLS instances but depends on the OS that is running on the client.

5.2.1. Configuring a Licensed Client on Linux with Default Settings

Perform this task from the client.

  1. As root, open the file /etc/nvidia/gridd.conf in a plain-text editor, such as vi.
    $ sudo vi /etc/nvidia/gridd.conf
    Note: You can create the /etc/nvidia/gridd.conf file by copying the supplied template file /etc/nvidia/gridd.conf.template.
  2. Add the FeatureType configuration parameter to the file /etc/nvidia/gridd.conf on a new line as FeatureType="value".

    value depends on the type of the GPU assigned to the licensed client that you are configuring.

    GPU Type Value
    NVIDIA vGPU 1. NVIDIA AI Enterprise automatically selects the correct type of license based on the vGPU type.
    Physical GPU The feature type of a GPU in pass-through mode or a bare-metal deployment:
    • 0: NVIDIA Virtual Applications
    • 2: NVIDIA RTX Virtual Workstation
    • 4: NVIDIA Virtual Compute Server
    Note: You can also perform this step from NVIDIA X Server Settings. Before using NVIDIA X Server Settings to perform this step, ensure that this option has been enabled as explained in NVIDIA AI Enterprise Client Licensing User Guide.

    This example shows how to configure a licensed Linux client for .

    # /etc/nvidia/gridd.conf.template - Configuration file for NVIDIA Grid Daemon
    …
    # Description: Set Feature to be enabled
    # Data type: integer
    # Possible values:
    # 0 => for unlicensed state
    # 1 => for NVIDIA vGPU
    # 2 => for NVIDIA RTX Virtual Workstation
    # 4 => for NVIDIA Virtual Compute Server
    FeatureType=
    ...
  3. Copy the client configuration token to the /etc/nvidia/ClientConfigToken directory.
  4. Ensure that the file access modes of the client configuration token allow the owner to read, write, and execute the token, and the group and others only to read the token.
    1. Determine the current file access modes of the client configuration token.
      # ls -l client-configuration-token-directory
    2. If necessary, change the mode of the client configuration token to 744.
      # chmod 744 client-configuration-token-directory/client_configuration_token_*.tok
    client-configuration-token-directory
    The directory to which you copied the client configuration token in the previous step.
  5. Save your changes to the /etc/nvidia/gridd.conf file and close the file.
  6. Restart the nvidia-gridd service.

The NVIDIA service on the client should now automatically obtain a license from the CLS or DLS instance.

5.2.2. Verifying the NVIDIA AI Enterprise License Status of a Licensed Client

After configuring a client with an NVIDIA AI Enterprise license, verify the license status by displaying the licensed product name and status.

To verify the license status of a licensed client, run nvidia-smi with the –q or --query optionfrom the licensed client, not the hypervisor host. If the product is licensed, the expiration date is shown in the license status.

nvidia-smi -q
==============NVSMI LOG==============

Timestamp                                 : Wed Nov 23 10:52:59 2022
Driver Version                            : 525.60.06
CUDA Version                              : 12.0

Attached GPUs                             : 2
GPU 00000000:02:03.0
    Product Name                          : 
    Product Brand                         : 
    Product Architecture                  : Ampere
    Display Mode                          : Enabled
    Display Active                        : Disabled
    Persistence Mode                      : Enabled
    MIG Mode
        Current                           : Disabled
        Pending                           : Disabled
    Accounting Mode                       : Disabled
    Accounting Mode Buffer Size           : 4000
    Driver Model
        Current                           : N/A
        Pending                           : N/A
    Serial Number                         : N/A
    GPU UUID                              : GPU-ba5b1e9b-1dd3-11b2-be4f-98ef552f4216
    Minor Number                          : 0
    VBIOS Version                         : 00.00.00.00.00
    MultiGPU Board                        : No
    Board ID                              : 0x203
    Board Part Number                     : N/A
    GPU Part Number                       : 25B6-890-A1
    Module ID                             : N/A
    Inforom Version
        Image Version                     : N/A
        OEM Object                        : N/A
        ECC Object                        : N/A
        Power Management Object           : N/A
    GPU Operation Mode
        Current                           : N/A
        Pending                           : N/A
    GSP Firmware Version                  : N/A
    GPU Virtualization Mode
        Virtualization Mode               : VGPU
        Host VGPU Mode                    : N/A
    vGPU Software Licensed Product
        Product Name                      : 
        License Status                    : Licensed (Expiry: 2022-11-23 10:41:16 GMT)
    …
    …

5.3. Installing NVIDIA Container Toolkit

Use NVIDIA Container Toolkit to build and run GPU accelerated Docker containers. The toolkit includes a container runtime library and utilities to configure containers to use NVIDIA GPUs automatically.



Ensure that the following software is installed in the guest VM:

Note: You do not need to install NVIDIA CUDA Toolkit on the hypervisor host.
  1. Set up the GPG key and configure apt to use NVIDIA Container Toolkit packages in the file /etc/apt/sources.list.d/nvidia-docker.list.
    $ distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
    $ curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
    $ curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
  2. Download information from all configured sources about the latest versions of the packages and install the nvidia-container-toolkit package.
    $ sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
  3. Restart the Docker service.
    $ sudo systemctl restart docker

5.4. Verifying the Installation of NVIDIA Container Toolkit

  1. Run the nvidia-smi command contained in the latest official NVIDIA CUDA Toolkit image.
    $ docker run --gpus all nvidia/cuda:11.0-base nvidia-smi
  2. Start a GPU-enabled container on any two available GPUs.
    $ docker run --gpus 2 nvidia/cuda:11.0-base nvidia-smi
  3. Start a GPU-enabled container on two specific GPUs identified by their index numbers.
    $ docker run --gpus '"device=1,2"' nvidia/cuda:10.0-base nvidia-smi
  4. Start a GPU-enabled container on two specific GPUs with one GPU identified by its UUID and the other GPU identified by its index number.
    $ docker run --gpus '"device=UUID-ABCDEF,1"' nvidia/cuda:11.0-base nvidia-smi
  5. Specify a GPU capability for the container.
    $ docker run --gpus all,capabilities=utility nvidia/cuda:11.0-base nvidia-smi

5.5. Installing Software Distributed as Container Images

The NGC container images accessed through the NVIDIA Enterprise Catalog includes the AI and data science applications, frameworks, and software in the infrastructure optimization and cloud native deployment layers. Each container image for an AI and data science application or framework contains the entire user-space software stack that is required to run the application or framework; namely, the CUDA libraries, cuDNN, any required Magnum IO components, TensorRT, and the framework.

Ensure that you have completed the following tasks in NGC Private Registry User Guide:

Perform this task from the VM.

For each AI or data science application that you are interested in, load the container as explained in Uploading an NVIDIA Container Image onto Your System in NGC Private Registry User Guide.

The following table lists the Docker pull command for downloading the container for each application or framework.

Application or Framework Docker pull Command
NVIDIA TensorRT
docker pull nvcr.io/nvaie/tensorrt-3-1:23.03-nvaie-3.1-py3
NVIDIA Triton Inference Server
docker pull nvcr.io/nvaie/tritonserver-3-1:23.03-nvaie-3.1-py3-sdk
NVIDIA Triton Inference Server
docker pull nvcr.io/nvaie/tritonserver-3-1:23.03-nvaie-3.1-py3-min
NVIDIA Triton Inference Server
docker pull nvcr.io/nvaie/tritonserver-3-1:23.03-nvaie-3.1-py3
PyTorch
docker pull nvcr.io/nvaie/pytorch-3-1:23.03-nvaie-3.1-py3
RAPIDS
docker pull nvcr.io/nvaie/nvidia-rapids-3-1:23.02-runtime-cuda12.1-ubuntu20.04
NVIDIA Clara Parabricks
docker pull nvcr.io/nvaie/clara-parabricks-3-1:4.0.3-1
NVIDIA DeepStream
docker pull nvcr.io/nvaie/deepstream-3-1:6.2-triton_nvaie
MONAI - Medical Open Network for Artificial Intelligence
docker pull nvcr.io/nvaie/monai-toolkit-3-1:1.0.1-1
TensorFlow 1
docker pull nvcr.io/nvaie/tensorflow-3-1:23.03-tf1-nvaie-3.1-py3
TensorFlow 2
docker pull nvcr.io/nvaie/tensorflow-3-1:23.03-tf2-nvaie-3.1-py3

The following table lists the Docker pull commands for downloading other software that is distributed as NGC container images through the NVIDIA Enterprise Catalog.

Other Software Docker pull Command
GPU Operator
docker pull nvcr.io/nvaie/gpu-operator-3-1:v23.3.1
Network Operator
docker pull nvcr.io/nvaie/network-operator-3-1:v23.1.0
vGPU Guest Driver, Ubuntu 22.04
docker pull nvcr.io/nvaie/vgpu-guest-driver-3-1:525.105.14-ubuntu22.04

5.6. Running ResNet-50 with TensorRT

  1. Launch the NVIDIA TensorRT container image on all GPUs in interactive mode, specifying that the container will be deleted when it is stopped.
    $ sudo docker run --gpus all -it --rm nvcr.io/nvaie/tensorrt:21.07-py3
  2. From within the container runtime, change to the directory that contains test data for the ResNet-50 convolutional neural network.
    # cd /workspace/tensorrt/data/resnet50
  3. Run the ResNet-50 convolutional neural network with FP32, FP16, and INT8 precision and confirm that each test is completed with the result PASSED.
    1. To run ResNet-50 with the default FP32 precision, run this command:
      # trtexec --duration=90 --workspace=1024 --percentile=99 --avgRuns=100 \
      --deploy=ResNet50_N2.prototxt --batch=1 --output=prob
    2. To run ResNet-50 with FP16 precision, add the --fp16 option:
      # trtexec --duration=90 --workspace=1024 --percentile=99 --avgRuns=100 \
      --deploy=ResNet50_N2.prototxt --batch=1 --output=prob --fp16
    3. To run ResNet-50 with INT8 precision, add the --int8 option:
      # trtexec --duration=90 --workspace=1024 --percentile=99 --avgRuns=100 \
      --deploy=ResNet50_N2.prototxt --batch=1 --output=prob --int8
  4. Press Ctrl+P, Ctrl+Q to exit the container runtime and return to the Linux command shell.

5.7. Running ResNet-50 with TensorFlow

  1. Launch the TensorFlow 1 container image on all GPUs in interactive mode, specifying that the container will be deleted when it is stopped.
    $ sudo docker run --gpus all -it --rm \
    nvcr.io/nvaie/tensorflow:21.07-tf1-py3
  2. From within the container runtime, change to the directory that contains test data for cnn example.
    # cd /workspace/nvidia-examples/cnn
  3. Run the ResNet-50 training test with FP16 precision.
    # python resnet.py --layers 50 -b 64 -i 200 -u batch --precision fp16
  4. Confirm that all operations on the application are performed correctly and that a set of results is reported when the test is completed.
  5. Press Ctrl+P, Ctrl+Q to exit the container runtime and return to the Linux command shell.

6. Additional Information

Additional information about containers and resources that are available from the NVIDIA AI Enterprise private registry on NGC is available in the documentation for these entities.

Container or Resource Additional Information
TensorFlow
PyTorch PyTorch Release Notes
NVIDIA Triton Inference Server Triton Inference Server Documentation on Github
NVIDIA TensorRT NVIDIA TensorRT Documentation
NVIDIA RAPIDS RAPIDS Docs on the RAPIDS project site
NVIDIA RAPIDS Accelerator for Apache Spark RAPIDS Accelerator for Apache Spark Deployment Guide
NVIDIA Clara Parabricks NVIDIA Clara Parabricks Documentation
NVIDIA DeepStream
MONAI (Medical Open Network for Artificial Intelligence) Enterprise NVIDIA MONAI Toolkit Documentation
TAO Toolkit TAO Toolkit Documentation
NVIDIA GPU Operator NVIDIA GPU Operator Documentation
NVIDIA Network Operator NVIDIA Network Operator Documentation

Notices

Notice

This document is provided for information purposes only and shall not be regarded as a warranty of a certain functionality, condition, or quality of a product. NVIDIA Corporation (“NVIDIA”) makes no representations or warranties, expressed or implied, as to the accuracy or completeness of the information contained in this document and assumes no responsibility for any errors contained herein. NVIDIA shall have no liability for the consequences or use of such information or for any infringement of patents or other rights of third parties that may result from its use. This document is not a commitment to develop, release, or deliver any Material (defined below), code, or functionality.

NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and any other changes to this document, at any time without notice.

Customer should obtain the latest relevant information before placing orders and should verify that such information is current and complete.

NVIDIA products are sold subject to the NVIDIA standard terms and conditions of sale supplied at the time of order acknowledgement, unless otherwise agreed in an individual sales agreement signed by authorized representatives of NVIDIA and customer (“Terms of Sale”). NVIDIA hereby expressly objects to applying any customer general terms and conditions with regards to the purchase of the NVIDIA product referenced in this document. No contractual obligations are formed either directly or indirectly by this document.

NVIDIA products are not designed, authorized, or warranted to be suitable for use in medical, military, aircraft, space, or life support equipment, nor in applications where failure or malfunction of the NVIDIA product can reasonably be expected to result in personal injury, death, or property or environmental damage. NVIDIA accepts no liability for inclusion and/or use of NVIDIA products in such equipment or applications and therefore such inclusion and/or use is at customer’s own risk.

NVIDIA makes no representation or warranty that products based on this document will be suitable for any specified use. Testing of all parameters of each product is not necessarily performed by NVIDIA. It is customer’s sole responsibility to evaluate and determine the applicability of any information contained in this document, ensure the product is suitable and fit for the application planned by customer, and perform the necessary testing for the application in order to avoid a default of the application or the product. Weaknesses in customer’s product designs may affect the quality and reliability of the NVIDIA product and may result in additional or different conditions and/or requirements beyond those contained in this document. NVIDIA accepts no liability related to any default, damage, costs, or problem which may be based on or attributable to: (i) the use of the NVIDIA product in any manner that is contrary to this document or (ii) customer product designs.

No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property right under this document. Information published by NVIDIA regarding third-party products or services does not constitute a license from NVIDIA to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property rights of the third party, or a license from NVIDIA under the patents or other intellectual property rights of NVIDIA.

Reproduction of information in this document is permissible only if approved in advance by NVIDIA in writing, reproduced without alteration and in full compliance with all applicable export laws and regulations, and accompanied by all associated conditions, limitations, and notices.

THIS DOCUMENT AND ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. TO THE EXTENT NOT PROHIBITED BY LAW, IN NO EVENT WILL NVIDIA BE LIABLE FOR ANY DAMAGES, INCLUDING WITHOUT LIMITATION ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, HOWEVER CAUSED AND REGARDLESS OF THE THEORY OF LIABILITY, ARISING OUT OF ANY USE OF THIS DOCUMENT, EVEN IF NVIDIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Notwithstanding any damages that customer might incur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described herein shall be limited in accordance with the Terms of Sale for the product.

VESA DisplayPort

DisplayPort and DisplayPort Compliance Logo, DisplayPort Compliance Logo for Dual-mode Sources, and DisplayPort Compliance Logo for Active Cables are trademarks owned by the Video Electronics Standards Association in the United States and other countries.

HDMI

HDMI, the HDMI logo, and High-Definition Multimedia Interface are trademarks or registered trademarks of HDMI Licensing LLC.

OpenCL

OpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.

Trademarks

NVIDIA, the NVIDIA logo, NVIDIA Maxwell, NVIDIA Pascal, NVIDIA Turing, NVIDIA Volta, Quadro, and Tesla are trademarks or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated.