Skip to content

Chapter 1: Operating Systems, Management Platforms, and First Power-On

HPC Networking FoundationsBeginner35 min read
Building the model…

The same cluster with all four networks drawn: compute rails, frontend, storage and out-of-band, each with its own switches and its own failure domain.

Chapter 0 showed you what the hardware looks like. Now you need to understand what runs on it, how you reach it, and what happens the first time you power it all on. This is the operational context that every CLI command in later chapters assumes you already have.

A senior engineer walking into an AI data centre for the first time knows exactly what questions to ask: What OS is on that switch? Does UFM manage the fabric or is it manually configured? Is there a console cable on these nodes or do I need IPMI? Can I make changes from the CLI or does the orchestration layer own the config?

This chapter answers those questions.


Read aloudPage 1