Hypervisors: type 1 versus type 2
In short: A hypervisor shares one physical computer between several virtual machines. Type 1 hypervisors run directly on the hardware and power data centres; type 2 hypervisors run as an application on a normal operating system and suit labs and desktops. Modern CPUs include features that make both efficient.
What a hypervisor does
A hypervisor, also called a virtual machine monitor, is software that creates and runs virtual machines (VMs). Each VM believes it has its own CPU, memory, disk and network card. The hypervisor schedules the real CPU between VMs, gives each a slice of memory, and keeps them isolated so a crash or compromise in one does not affect the others. For the wider idea, see virtualisation basics.
Type 1: bare-metal
A type 1 hypervisor is installed directly on the server instead of a normal operating system. Examples include VMware ESXi, Microsoft Hyper-V, KVM (built into the Linux kernel, used by Proxmox VE and many clouds) and Xen. Because there is no general-purpose OS underneath, overhead is low and there is less code for an attacker to target. Type 1 hypervisors are managed remotely, often through a central console such as VMware vCenter, System Center or the Proxmox web interface.
Hyper-V is sometimes confused for type 2 because you add it from Windows. In fact, once enabled, Hyper-V sits beneath Windows, and Windows itself becomes a special privileged partition. KVM similarly turns Linux into a hypervisor.
Type 2: hosted
A type 2 hypervisor runs as an application on an existing operating system. Examples are Oracle VirtualBox, VMware Workstation and VMware Fusion on macOS. It is easy to install and ideal for learning, testing software and running a home lab on a laptop. The trade-off is that every request passes through the host OS, which adds overhead, and a problem with the host affects every VM.
| Scenario | Sensible choice | Why |
|---|---|---|
| Student lab on a laptop | Type 2 (VirtualBox, Workstation) | Runs alongside your normal desktop |
| Home lab on a spare PC | Type 1 (Proxmox VE, Hyper-V Server role) | Better performance, closer to real work |
| Company data centre | Type 1 with central management | High availability, live migration, scale |
| Public cloud | Provider's type 1 (often KVM-based) | You rent VMs; the provider runs the hypervisor |
Hardware support and practical tips
Modern CPUs include virtualisation extensions: Intel VT-x and AMD-V for running guest code efficiently, plus EPT/NPT (second-level address translation) for fast memory mapping. VT-d or AMD-Vi (IOMMU) allows a physical device such as a GPU or NIC to be passed directly to one VM. These are often disabled in the firmware by default, so enable them in the BIOS or UEFI before installing a hypervisor.
Overcommit means giving VMs more virtual CPUs or memory in total than the host physically has, on the basis that they will not all be busy at once. It improves density but can cause slowdowns if pushed too far. Watch CPU ready time and memory ballooning or swapping as warning signs.
Tip: licensing and vendor changes affect hypervisor choice. Broadcom's changes to VMware licensing after its 2023 acquisition led many organisations to review alternatives, so check current terms rather than relying on old assumptions.
Practice
Is VirtualBox a type 1 or type 2 hypervisor?
Why do type 1 hypervisors usually have a smaller attack surface?
A hypervisor will not start any VMs on a new PC. What firmware setting should you check?
Key takeaways
- A hypervisor shares hardware between isolated virtual machines.
- Type 1 runs on bare metal: ESXi, Hyper-V, KVM, Xen.
- Type 2 runs on a host OS: VirtualBox, VMware Workstation.
- Enable VT-x or AMD-V (and IOMMU for passthrough) in firmware.
- Overcommit carefully and monitor for contention.
Original content written by Mohamed Fayas for FM TecHub. © 2026 FM TecHub. You're welcome to share a link to this page; please don't copy the text elsewhere without permission.