To enable developers to deploy AI/ML workloads on TKG clusters, as a vSphere Administrator you set up the Supervisor environment to support NVIDIA GPU hardware.

Admin Step 1: Review System Requirements

Refer to the following system requirements to set up the environment for deploying AI/ML workloads on TKG clusters.
Requirement Description

vSphere 8 infrastructure

vCenter Server and ESXi hosts

Workload Management license

vSphere Namespaces and Supervisor
TKR Ubuntu OVA

Tanzu Kubernetes Releases Release Notes

NVIDIA vGPU Host Driver

Download the VIB from the NGC web site. For more information, see the vGPU Software Driver documentation.

NVIDIA License Server for vGPU

FQDN provided by your organization

Admin Step 2: Install Supported NVIDIA GPU Device on ESXi Hosts

To deploy AI/ML workloads on TKG, you install one or more supported NVIDIA GPU devices on each ESXi host comprising the vCenter Cluster where Workload Management will be enabled.

To view compatible NVIDIA GPU devices, refer to the VMware Compatibility Guide.

The list of compatible NVIDIA GPU devices. Click on a GPU device model to view more details and to subscribe to RSS feeds.

The NVIDA GPU device should support the latest NVIDIA AI Enterprise (NVAIE) vGPU profiles. Refer to the NVIDIA Virtual GPU Software Supported GPUs documentation for guidance.

For example, the following ESXi host has two NVIDIA GPU A100 devices installed on it.

The Graphics Devices tab in the vSphere Client lists the NVIDIA GPU A100 devices.

Admin Step 3: Configure Each ESXi Host for vGPU Operations

For each ESXi host comprising the vCenter Cluster where Workload Management is enabled, configure the host for NVIDIA vGPU by enabling Shared Direct and SR-IOV.

Enable Shared Direct on Each ESXi Host

For NVIDIA vGPU functionality to be unlocked, enable Shared Direct mode on each ESXi host comprising the vCenter Cluster where Workload Management is enabled.

To enable Shared Direct, complete the following steps. For additional guidance, see Configure Virtual Graphics on vSphere.
  1. Log on to the vCenter Server using the vSphere Client.
  2. Select an ESXi host in the vCenter Cluster.
  3. Select Configure > Hardware > Graphics > Graphics Devices.
  4. Select the NVIDIA GPU accelerator device.
  5. Edit the Graphics Device settings.
  6. Select Shared Direct.
  7. For Shared passthrough GPU assignment policy, for best performance select Spread VMs across GPUs
  8. Click OK to save the configuration.
  9. Note that the settings will take effect after restarting the host.
  10. Right-click the ESXi host and put it into maintenance mode.
  11. Reboot the host.
  12. When the host is running again, take it out of maintenance mode.
  13. Repeat this process for each ESXi host in the vSphere cluster supporting Workload Management .

Turn On SR-IOV BIOS for NVIDIA GPU A30 and A100 Devices

If you are using the NVIDIA A30 or A100 GPU devices, which are required for Multi-Instance GPU (MIG mode), you must enable SR-IOV on the ESXi host. If SR-IOV is not enabled, Tanzu Kubernetes cluster node VMs cannot start. If this occurs, you see the following error message in the Recent Tasks pane of the vCenter Server where Workload Management is enabled.

Could not initialize plugin libnvidia-vgx.so for vGPU nvidia_aXXX-xx. Failed to start the virtual machine. Module DevicePowerOn power on failed.

To enable SR-IOV, log in to the ESXi host using the web console. Select Manage > Hardware . Select the NVIDIA GPU device and click Configure SR-IOV. From here you can turn on SR-IOV. For additional guidance, see Single Root I/O Virtualization (SR-IOV) in the vSphere documentation.

vGPU with Dynamic DirectPath IO (Passthrough Enabled Device)

If you are using vGPU with Dynamic DirectPath IO, perform the following additional configuration.
  1. Log in to the vCenter Server using the vSphere Client.
  2. Select the target ESXi host in the vCenter Cluster.
  3. Select Configure > Hardware > PCI Devices.
  4. Select the All PCI Devices tab.
  5. Select the target NVIDIA GPU accelerator device.
  6. Click Toggle Passthrough.
  7. Right-click the ESXi host and put it into maintenance mode.
  8. Reboot the host.
  9. When the host is running again, take it out of maintenance mode.

Admin Step 4: Install the NVIDIA Host Manager Driver on Each ESXi Host

To run Tanzu Kubernetes cluster node VMs with NVIDIA vGPU graphics acceleration, you install the NVIDIA host manager driver on each ESXi host comprising the vCenter Cluster where Workload Management will be enabled.

The NVIDIA vGPU host manager driver components are packaged in a vSphere installation bundle (VIB). The NVAIE VIB is provided to you by your organization through its NVIDIA GRID licensing program. VMware does not provide NVAIE VIBs or make them available for download. As part of the NVIDIA licensing program your organization sets up a licensing server. Refer to the NVIDIA Virtual GPU Software Quick Start Guide for more information.

Once the NVIDIA environment is set up, run the following command on each ESXi host, replace the NVIDIA license server address and the NVAIE VIB version and with the appropriate values for your environment. For additional guidance, see Installing and configuring the NVIDIA VIB on ESXi at the VMware Support Knowledge Base.
Note: The NVAIE VIB version installed on ESXi hosts must match the vGPU software version installed the node VMs. The version below is only an example. 
esxcli system maintenanceMode set --enable true
esxcli software vib install -v ftp://server.domain.example.com/nvidia/signed/NVIDIA_bootbank_NVIDIA-VMware_ESXi_7.0_Host_Driver_460.73.02-1OEM.700.0.0.15525992.vib
esxcli system maintenanceMode set --enable false
/etc/init.d/xorg restart

Admin Step 5: Verify ESXi Hosts Are Ready for NVIDIA vGPU Operations

To verify that each ESXi host is ready for NVIDIA vGPU operations, perform the following checks on each ESXi host in the vCenter Cluster where Workload Management will be enabled:
  • SSH into the ESXi host, enter shell mode and run the command nvidia-smi. The NVIDIA System Management Interface is a command line utility provided by the NVIDA vGPU host manager. Running this command returns the GPUs and drivers on the host.
  • Run the following command to verify that the NVIDIA driver is properly installed: esxcli software vib list | grep NVIDA.
  • Verify that host is configured with GPU shared direct and that SR-IOV is turned on (if you are using NVIDIA A30 or A100 devices).
  • Using the vSphere Client, on the ESXi host that is configured for GPU, create a new virtual machine with a PCI device included. The NVIDIA vGPU profile should appear and be selectable.The Customize Hardware tab with NVIDIA vGPU profile selected.

Admin Step 6: Enable Workload Management

To enable Workload Management, see Deploying TKG Service Clusters.
Note: Skip this step if you already have a vSphere Cluster with Workload Management enabled, assuming the cluster is using the ESXi hosts you have configured for vGPU.

Admin Step 7: Create or Update a Content Library with a TKR Ubuntu

NVIDIA vGPU requires the Ubuntu operating system. You cannot use the PhotonOS edition of a Tanzu Kubernetes release for vGPU clusters.

VMware provides an Ubuntu editions of Tanzu Kubernetes releases. Starting with vSphere 8, the Ubuntu edition is specified using an annotation in the cluster YAML.

Create or update an existing content library with a supported Ubuntu TKR. See Administering Kubernetes Releases for TKG Service Clusters.
Note: Skip this step if you already have an existing TKR content library configured on vCenter. Do not create a second content library for TKRs. Doing so can cause system instability.

Admin Step 8: Create a Custom VM Class with the vGPU Profile

Create a custom VM Class with a vGPU profile. You will then use this VM class in the cluster specification to create the TKGS cluster nodes. Refer to the following instructions: Create a Custom VM Class for NVIDIA vGPU Devices.

Admin Step 9: Configure the vSphere Namespace

Create a vSphere Namespace for each TKG vGPU cluster you plan to provision. See Create a vSphere Namespace for Hosting TKG Service Clusters.

Configure the vSphere Namespace by adding vSphere SSO users/groups with Edit permissions, and attach a storage policy for persistent volumes. See Configure a vSphere Namespace for TKG Service Clusters.

Associate the TKR content library where the desired Ubuntu image is stored with the vSphere Namespace. See Associate the TKR Content Library with the TKG Service.

Associate the custom VM class with the vSphere Namespace.
  • In the Select the vSphere Namespace, select the VM Service tile and click Manage VM Classes.
  • Locate the custom VM class you created in the list of classes.
  • Select (check) the class and click Add.
For additional guidance, see Associate VM Classes with the vSphere Namespace.

Admin Step 10: Verify that Supervisor Is Ready

The last administration task is to verify that Supervisor is provisioned and available for use by the Cluster Operator to provision a TKG cluster for AI/ML workloads.

See Connecting to TKG Service Clusters Using vCenter SSO Authentication.