Implementing VMware Private AI Foundation with NVIDIA on top of the private AI infrastructure components includes creating vector databases by using VMware Data Services Manager, and deploying VMs for AI development based on NVIDIA DL workloads images and GPU-enabled Tanzu Kubernetes Grid (TKG) clusters for running NVIDIA NGC container images.

You can deploy and configure VMware Private AI Foundation with NVIDIA according to two implementation models - a cloud-connected or a disconnected environment.

For information on the VMware Private AI Foundation with NVIDIA design, see Detailed Design for VMware Private AI Foundation with NVIDIA for Private AI Ready Infrastructure for for VMware Cloud Foundation.

Prerequisites

To complete the implementation of VMware Private AI Foundation with NVIDIA for Private AI Ready Infrastructure for VMware Cloud Foundation validated solution, verify that your system fulfills the following prerequisites.

Table 1. Prerequisites for Adding VMware Private AI Foundation with NVIDIA to Private AI Ready Infrastructure for VMware Cloud Foundation

Category

Prerequisite

Environment

NVIDIA NGC API key

Verify that you have an API key for access to the nvcr.io private registry.

Docker client Verify that your system has a machine with Docker installed.
VMware Tanzu Network account Verify that you have an account for VMware Tanzu Network so that you can download VMware Data Services Manager software.