Overview#

This Deployment Guide for NVIDIA Enterprise RAs – Base Command Manager provides a step-by-step procedure for installing NVIDIA Base Command Manager (BCM) on bare-metal cluster hardware. The instructions focus on concise, practical steps with minimal explanations so that a moderately experienced cluster administrator can get a cluster up and running in a standard configuration as quickly as possible. Base Command Manager is one of several orchestration options available for data center cluster management.

Base Command Manager is included in NVIDIA AI Enterprise, which brings together optimized microservices, frameworks and libraries for AI development with advanced GPU orchestration and infrastructure management into a production-grade software suite to power your business

It is strongly recommended that the cluster administrator be familiar with the NVIDIA Enterprise Reference Architecture document sets and the components that form their design and implementation. The steps are intended to be hardware-agnostic, meaning they can be adapted for hardware configurations different from those illustrated with minimal changes. Be sure to replace all example values with actual information relevant to the target environment, especially:

  • Cluster specific information such as names, locations and passwords

  • IP addresses, including name servers and gateways

  • Interface names

Information provided in code blocks reflects the context in which each command is executed. In most cases, these commands are run from within cmsh (Cluster Management Shell):

[nvidia-era-bcm-01]% device

However, some commands may be run as the root user directly from within the operating system:

root@nvidia-era-bcm-01:~# cmsh

For more detailed information, please refer to Appendix A: Related Links.