About the VMware Private AI Foundation with NVIDIA Guide

The VMware Private AI Foundation with NVIDIA Guide provides an overview of the components of VMware Private AI Foundation with NVIDIA and high-level workflows for development and production use cases.

Intended Audience

The information in VMware Private AI Foundation with NVIDIA Guide is intended for data center cloud administrators, data scientists, and DevOps engineers who are familiar with:

• Cloud administrators
  • Concepts of virtualization and software-defined data centers (SDDCs)
  • Hardware components such as top-of-rack (ToR) switches, inter-rack switches, servers with direct attached storage, cables, and power supplies
  • Methods for setting up NVIDIA GPUs on servers in a data center
  • Using VMware vSphere^® to work with virtual machines.
  • Using vSphere with Tanzu to configure and assign vSphere resources to vSphere namespaces on a Supervisor.
• Data scientists
  • Provisioning virtual machines in vSphere with Tanzu by using the VM Service and the Kubernetes API.
  • Containers, including Docker, Helm charts and Harbor Registry
• DevOps engineers
  • Provisioning virtual machines in vSphere using the Kubernetes API.
  • Containers, including Docker, Helm charts and Harbor Registry
  • Working with vSphere with Tanzu for provisioning VMs and Tanzu Kubernetes Grid (TKG) clusters.

VMware Software Components

The functionality of the VMware Private AI Foundation with NVIDIA solution is available across several software components.

Software Category	Supported Software Versions
Management components	See VMware Components in VMware Private AI Foundation with NVIDIA.
Deep learning VM components	See VMware Deep Learning VM Release Notes.
TK releases (TKr)	See VMware Tanzu Kuberenetes releases Release Notes.

Related VMware Documentation

The VMware Private AI Foundation with NVIDIA solution includes a stack of VMware software products and components. The documentation for those software products is as follows:

• VMware Cloud Foundation Documentation
• VMware vSphere and vSAN Documentation
• VMware vSphere with Tanzu Documentation
• VMware Aria Automation Documentation
• VMware Aria Operations Documentation
• VMware Aria Suite Lifeycle Documentation
• VMware Data Services Manager Documentation

VMware Cloud Foundation Glossary

The VMware Cloud Foundation Glossary defines terms specific to VMware Cloud Foundation.

Update History

This VMware Private AI Foundation with NVIDIA Guide is updated when necessary.

Revision	Description
14 MAY 2024	You can now use VMware Aria Automation 8.17 for provisioning AI workloads in VMware Private AI Foundation with NVIDIA. See Set Up VMware Aria Automation for VMware Private AI Foundation with NVIDIA.
06 MAY 2024	You can now use VMware Aria Operations 8.17.1 for monitoring workload domains running VMware Private AI Foundation with NVIDIA. See Monitoring VMware Private AI Foundation with NVIDIA. The documentation is updated for the latest versions of the deep learning VM images. See Deploying a Deep Learning VM in VMware Private AI Foundation with NVIDIA and Deploying RAG Workloads in VMware Private AI Foundation with NVIDIA. Learn more about deploying a deep learning VM with a static IP address. See Assign a Static IP Address to a Deep Learning VM in VMware Private AI Foundation with NVIDIA. For a disconnected environment, to easily keep the NVIDIA NGC container images in the local Harbor registry up-to-date, use a setup with two registry instances. See Setting Up a Private Harbor Registry in VMware Private AI Foundation with NVIDIA. For efficient metrics collection, the documentation is updated with details about running a deep learning (DL) workload and DCGM Exporter in the same deep learning VM. See DCGM Exporter. Learn more about deploying a database in VMware Data Services Manager by using a self-service catalog item in VMware Aria Automation. See Deploying a Vector Database by Using a Self-Service Catalog Item in VMware Aria Automation.
26 MAR 2024	Initial version.