What is Hypervisor?
Every virtual machine has a foundation. Meet the hypervisor.
Understand the software behind virtual machines, the difference between Type 1 and Type 2 hypervisors, and how familiar platforms put virtualization to work.
What is a hypervisor?
Imagine a company owns one powerful physical server but needs to run an accounting application, a customer portal and a testing environment. Installing everything inside one operating system could create conflicts and make changes difficult. Buying three separate servers is another option, but may leave expensive hardware underused.
A hypervisor offers a third approach. It divides the server’s available resources into multiple virtual computers, each with its own operating system and applications. To the software running inside, each VM appears to have a dedicated processor, memory, disk and network adapter. Underneath, those components are supplied by shared physical hardware.
This is the foundation of server virtualization. The hypervisor is sometimes called a virtual machine monitor (VMM). It is not a physical device or a complete data center: it is the virtualization technology that makes VMs possible.
How does a hypervisor work?
The hypervisor stands between virtual workloads and the physical resources they depend on. It creates virtual hardware, schedules access to processing power, coordinates memory use and sends storage or network requests through the host’s I/O stack.
Create a VM
An administrator defines a VM with virtual CPUs, memory, storage and networking, then installs a guest operating system.
Allocate resources
The hypervisor maps those virtual resources to the host’s CPUs, RAM, physical storage and network interfaces.
Run independently
The VM runs applications in its own environment while the hypervisor schedules and isolates competing workloads.
For example, a VM configured with four virtual CPUs does not necessarily own four dedicated physical CPU cores. The hypervisor can schedule those virtual CPUs alongside other VMs, depending on utilization, reservations and platform settings. Similarly, a virtual disk may be stored in a file, a logical storage volume or a distributed storage system.
Virtualization also depends on hardware features. Modern processors provide acceleration such as Intel VT-x or AMD-V, while compatible drivers and I/O technologies help the host work efficiently. A good design still requires enough actual capacity: virtualization does not create unlimited processing power or memory.
Type 1 vs. Type 2 hypervisors: what’s the difference?
Hypervisors are commonly divided into two categories based on where the virtualization layer sits. The distinction helps explain why a platform is suited to a production server, a developer’s laptop, or both.
Runs as an application above a general-purpose operating system, making it convenient for local testing, software development and running another desktop OS.
Common Type 1 examples: VMware ESXi, Nutanix AHV, Microsoft Hyper-V and KVM-based server virtualization systems. Type 2 examples: Oracle VirtualBox and VMware Workstation.
One useful nuance: KVM is built into the Linux kernel, so its design does not look exactly like a small standalone hypervisor. It is nevertheless generally classified as Type 1 because Linux, with KVM enabled, provides the virtualization host itself. The Type 1/Type 2 labels are useful shorthand, not a complete measure of security or performance.
What does a hypervisor actually manage?
It does more than start and stop virtual machines. These six capabilities determine much of the experience of operating virtualized servers.
CPU scheduling
Shares physical processor time among virtual CPUs while balancing workload demand and scheduling policies.
Memory management
Tracks assigned RAM and may support features such as ballooning or reclamation, depending on the platform.
Virtual storage
Presents virtual disks and coordinates I/O to physical disks, shared storage or software-defined storage layers.
Networking
Connects virtual network adapters to switches, networks and physical interfaces through a virtual networking stack.
Isolation
Helps separate guest operating systems so a problem in one VM does not normally alter another VM’s memory or execution.
VM lifecycle
Supports VM creation, power operations and other management functions, often exposed through a larger software platform.
Remember: High availability, live migration, distributed scheduling and backup are often delivered by the broader virtualization platform and its supporting infrastructure. They are not automatic properties of every hypervisor installation.
Hypervisor vs. VM vs. virtualization platform vs. HCI
These terms appear together in product descriptions, but they refer to different layers. Understanding the distinction makes vendor comparisons more meaningful.
Swipe sideways to read the full comparison →
| Term | What it means | Simple example |
|---|---|---|
| Hypervisor | The software component that runs and controls virtual machines. | VMware ESXi or KVM |
| Virtual machine | A software-defined computer with its own guest OS and allocated virtual hardware. | A Windows Server VM hosting an application |
| Virtualization platform | Hypervisor technology together with management, monitoring and other operational tools. | VMware vSphere or Proxmox VE |
| HCI | An integrated architecture bringing virtualized compute together with software-defined storage and infrastructure services. | Nutanix Cloud Infrastructure or Sangfor HCI |
Examples describe their general role; specific product bundles and features vary.
A hypervisor can operate without a complete hyperconverged infrastructure platform. Conversely, hyperconverged infrastructure (HCI) typically includes a hypervisor plus coordinated storage, networking and management. The added components matter when comparing operational effort and total cost.
Why are hypervisors so important?
Before server virtualization became widespread, individual applications were frequently tied to dedicated physical hardware. That approach could make maintenance, upgrades and capacity planning slow and expensive. Hypervisors helped organizations consolidate systems while keeping operating environments logically separate.
Use hardware more efficiently
Multiple VMs can share a server instead of leaving significant computing capacity idle.
Provision workloads faster
IT teams can deploy a new virtual server without ordering hardware for every application.
Separate workloads
Different operating systems and software stacks can coexist with defined resource and security boundaries.
Support infrastructure change
With the right platform and design, VMs can be relocated or recovered without rebuilding an application from scratch.
Those advantages come with responsibilities. A single host becomes an important shared dependency, so teams must plan hardware redundancy, patching, access control and tested recovery. Too many VMs competing for CPU, memory or storage can cause performance problems. Snapshots are useful for short-term changes, but should not be treated as a substitute for independently tested backups.
Hypervisor technologies: VMware, Nutanix, Proxmox and Sangfor
These four names illustrate different ways hypervisor technology is delivered and managed. They are not a performance ranking; each needs to be evaluated against the applications, support expectations and infrastructure in use.
VMware ESXi
ESXi is VMware’s bare-metal hypervisor and underpins its vSphere virtualization environment. Organizations often evaluate it for established enterprise workflows, mature VM operations and broad ecosystem compatibility. The surrounding edition and management tools determine the complete experience.
Explore VMware vSphere ↗Nutanix AHV
AHV is a KVM-based enterprise hypervisor closely integrated with Nutanix Cloud Infrastructure and Prism management. Its approach is built around operating VMs as part of a wider infrastructure platform, making full-stack requirements important in any evaluation.
Explore Nutanix AHV ↗Proxmox VE with KVM
Proxmox Virtual Environment uses KVM to run full virtual machines and LXC for Linux containers. Proxmox VE is the platform, while KVM is the hypervisor underneath its VMs. This model offers flexibility alongside responsibility for platform design and operations.
Explore Proxmox VE ↗Sangfor aSV
Sangfor aSV is a Type 1, KVM-based hypervisor that provides compute virtualization. It also serves as the foundation of Sangfor HCI, which adds software-defined storage, networking and security functions. This makes it relevant to teams considering either virtualization or a more integrated deployment.
Explore Sangfor aSV ↗The important distinction is not simply whether a product uses KVM or another underlying technology. Enterprise buyers should also examine the management experience, supported workloads, recovery ecosystem, update process and service commitments.
How to choose the right hypervisor
Start with the applications you need to run rather than a list of impressive features. A sensible shortlist should answer six questions before committing to a migration or new deployment.
- Compatibility: Are your guest operating systems, applications, virtual appliances, hardware and backup tools supported?
- Performance: Can the design deliver predictable CPU, memory, storage and network performance under realistic peak demand?
- Resilience: What happens when a host fails, a storage path disappears or an upgrade is interrupted?
- Manageability: How will administrators monitor VMs, apply updates, automate tasks and troubleshoot issues across hosts?
- Security and support: What controls, patch processes, technical assistance and response times are available?
- Long-term cost: What is the three- to five-year cost of software, hardware, migrations, support, training and operational effort?
Test before deciding
Build a small proof of concept with representative applications. Measure CPU contention, memory pressure, storage latency, backup restoration and VM behavior during a simulated host outage. A hypervisor that looks good in a feature matrix still needs to prove itself with your workloads.
Hypervisor FAQs
Is a hypervisor the same as a virtual machine?
No. The hypervisor is the technology that runs and manages virtual machines. A VM is the guest system created on top of that technology, with its own operating system and applications.
Is KVM a Type 1 hypervisor?
KVM is generally classified as Type 1 because it enables Linux to function as a virtualization host. It uses Linux kernel capabilities and other host components rather than operating as a separate standalone kernel.
Can a hypervisor run Windows and Linux at the same time?
Yes, when the hypervisor and host hardware support those guest operating systems. Each OS runs inside its own VM, with resources allocated and managed by the host.
Does a hypervisor automatically provide high availability?
No. Many virtualization platforms offer high-availability features, but they typically require multiple hosts, compatible storage and networking, appropriate licenses or subscriptions, and careful configuration.
Is a hypervisor necessary for containers?
Not always. Containers normally share an operating-system kernel and do not require a VM hypervisor. In practice, organizations often run containers inside virtual machines, combining both approaches.
The hypervisor is the engine. The platform is the complete experience.
Understanding hypervisors makes it easier to judge virtualization products beyond their labels. Start with the applications, look closely at how the technology allocates resources and supports operations, and evaluate the whole environment rather than a single component.
Compare virtualization platforms →Sources and further reading
Definitions are based on general virtualization references; vendor descriptions are based on official product documentation. Details can vary by version, deployment and support arrangement.