Skip to content

1. Virtualization Basics

A hypervisor lets one physical machine run several separate "computers" at once, each with its own OS, isolated from the others. Each of those is a virtual machine (VM).

This isn't hypothetical — it's the lab

pve1 and pve2 run Proxmox VE, which is itself a hypervisor (built on KVM, Linux's built-in virtualization). Every "server" you've touched in this course — your personal MUD, the lab desktop, the course wiki's own CT 251 — is a guest running on top of just two physical machines.

Two different kinds of guest, both in this lab

Virtual Machine (VM) Container (LXC)
What it virtualizes A whole computer, including its own kernel Just an isolated userspace — it shares the host's kernel
Can run a different OS than the host? Yes (e.g. Windows VM on a Linux host) No — same kernel as the host, always
Weight Heavier, slower to start Lighter, starts almost instantly
Example in this lab win-rds (VM 920) — a full Windows Server lab-guacamole (CT 251) — a Linux container

That's exactly why win-rds had to be a full VM and not a container: it runs Windows, a different kernel entirely from pve1's Linux host. A container physically cannot do that — it's not a different, lighter kind of VM, it's a fundamentally different technique that only works when the guest and host already share a kernel.

Try it

Without looking it up, guess whether CT 252 (lab-mud, running Evennia) is a VM or a container, and explain your reasoning the same way the table above does for win-rds.

Tip

"Type 1" hypervisors (like Proxmox/KVM) run directly on hardware with no underlying OS to speak of. "Type 2" hypervisors run as an application on top of a regular OS (like VirtualBox on your Windows laptop) — Lesson 2 is all about that second kind.

Play it

Open the Virtualization and Cloud Lab and use the VM vs container tab.

Next: Client-Side Virtualization →