In this design guide, we present guidelines for implementing an AI platform at enterprise scale based on the results produced in the Dell Technologies HPC & AI innovations Lab. Our lab setup included servers, storage, and networking from Dell Technologies, virtualization software from VMware, and acceleration, networking, and software from NVIDIA configured with the combined best practices from all three partners. This design guide describes all the recommended configurations, network topologies, deployment guidelines, and observed performance.
Using the information in this design guide, experienced VMware professionals can quickly use their existing skills to deploy and operate a full-featured platform to support advanced AI use cases.
This design guide is a companion guide to the Virtualizing GPUs for AI with VMware and NVIDIA technical white paper. For more details about the technology components used in this validated design, see the white paper.