Server Virtualization Glossary
One physical server. Many virtual machines.
Understand what server virtualization means, how a hypervisor works, and the essential terms behind modern IT infrastructure — without unnecessary jargon.
What is server virtualization?
Think of a physical server as an apartment building. The building supplies shared foundations, electricity and utilities; individual apartments provide separate spaces for different tenants. Server virtualization works in a similar way: one physical machine supplies processing power, memory, storage and network connectivity to several virtual servers.
Traditionally, a company might run its accounting system, internal website and test environment on three separate physical machines. With server virtualization, it may be able to run those workloads as separate VMs on one adequately sized server. Each VM behaves like an individual computer, even though all three share physical hardware.
This separation makes it easier to provision systems, change resource allocations and manage multiple workloads. It also introduces planning requirements: if the host fails, every VM depending on it may be affected unless suitable resilience is designed in.
How does server virtualization work?
The basic architecture has three layers: physical hardware, a hypervisor, and virtual machines. The hypervisor allocates resources and keeps VMs logically separated. Management software adds tools to monitor, automate and protect workloads across one or many servers.
Physical server
The host provides real CPU cores, RAM, disks and network connections. All VMs ultimately depend on these resources.
Hypervisor
Software creates virtual hardware for each VM, schedules CPU time and manages access to memory, storage and I/O.
Virtual machines
Each VM runs a guest operating system and its own applications, with defined virtual CPUs, memory, disks and networks.
For example, a host with sufficient capacity might run a Windows business application, a Linux web server and a separate test VM. They can be started, stopped and administered independently. Their assigned virtual resources still need to be managed against the host’s actual capacity.
What is a hypervisor?
A hypervisor is the software layer that creates and runs VMs. It presents virtual hardware to operating systems and coordinates their access to the physical host. Two broad types are commonly discussed.
Built for servers
Runs directly on host hardware, or as part of a purpose-built virtualization host stack. It is the usual model for enterprise data centers. Examples include VMware ESXi and KVM-based platforms.
Built for local machines
Runs as an application on a conventional desktop operating system. It is often useful for development, training and testing; Oracle VirtualBox is a familiar example.
Many enterprise products go beyond the hypervisor itself. Their management, clustering, networking and storage functions are what turn individual virtualization hosts into an operational platform.
Why do organizations use server virtualization?
Better hardware utilization
Multiple workloads can share server capacity that might otherwise remain underused.
Faster provisioning
IT teams can deploy a new VM from a template without purchasing a new physical machine for every application.
Flexible operations
VMs can be resized, cloned or moved between compatible hosts when the platform and design allow it.
Potential cost savings
Consolidation may lower hardware, rack space, power and cooling costs, although software and support expenses still count.
Important: Virtualization is not automatic high availability, and a VM snapshot is not a complete backup strategy. Teams still need capacity planning, security controls, tested backups and recovery procedures.
Virtual machines vs. containers: what’s the difference?
A virtual machine includes its own guest operating system, while a container packages an application and its dependencies using isolation provided by a shared operating-system kernel. Both are useful, but they solve different problems.
Swipe sideways to see the full table →
| Consideration | Virtual machines | Containers |
|---|---|---|
| Operating system | Separate guest OS inside each VM | Share the host kernel |
| Typical footprint | Usually larger because an OS is included | Usually lighter and faster to start |
| Good for | Legacy applications, mixed operating systems and strong OS boundaries | Portable applications, microservices and rapid deployment |
| Can they work together? | Yes. Containers frequently run inside VMs, combining application flexibility with infrastructure isolation. | |
Is server virtualization the same as HCI?
No. Server virtualization creates and manages VMs by abstracting compute resources. Hyperconverged infrastructure (HCI) combines virtualization with software-defined storage and integrated infrastructure management, and typically virtual networking, in a coordinated platform.
Think of it this way
The hypervisor runs the virtual machines. An HCI platform provides a more complete foundation around them, coordinating compute, storage and other infrastructure services. An organization can virtualize servers without adopting HCI, while most HCI platforms rely on virtualization at their core.
Common server virtualization platforms
Virtualization platforms differ in their approach to licensing, operations, storage integration and scale. Four commonly evaluated names illustrate the range of choices; they are examples rather than a ranked list.
VMware vSphere
Built around VMware ESXi and enterprise management capabilities, vSphere is widely used to run and operate business workloads in virtual machines.
Explore VMware vSphere ↗Nutanix AHV
AHV provides enterprise virtualization within the Nutanix infrastructure portfolio, supported by unified management through Prism.
Explore Nutanix AHV ↗Proxmox VE
Proxmox Virtual Environment manages KVM-based VMs and LXC containers through a web interface, with clustering and storage integration.
Explore Proxmox VE ↗Sangfor aSV & HCI
Sangfor aSV is a KVM-based hypervisor. Sangfor HCI builds on it by integrating virtualization with software-defined storage, networking and security functions.
Explore Sangfor’s infrastructure stack ↗Rather than choosing by brand alone, assess application support, performance, operational skills, recovery requirements, contract terms and the total cost of running your environment.
Server virtualization glossary: 16 essential terms
These are the words you’ll encounter most often when reading specifications, evaluating hypervisors or planning a virtualization project.
1. Hypervisor
The software that creates VMs and controls how they use physical hardware resources.
2. Host
The physical computer that supplies CPU, RAM, storage and networking to virtual machines.
3. Virtual machine (VM)
A software-defined computer that runs an operating system and applications on shared hardware.
4. Guest OS
The operating system installed inside a virtual machine, such as Windows Server or Linux.
5. vCPU
A virtual processor assigned to a VM; its execution is scheduled on the host’s physical CPUs.
6. VM memory (vRAM)
RAM capacity assigned to a VM and managed against the physical host’s available memory.
7. Virtual disk
Logical storage presented to a VM, backed by physical or software-defined storage.
8. Virtual switch (vSwitch)
Software that connects VMs to each other and, when configured, to physical networks.
9. VM template
A prepared virtual machine image or configuration used to create consistent new VMs.
10. Snapshot
A point-in-time representation of a VM’s state or disk data, useful for short-term operations but not a substitute for backups.
11. High availability (HA)
Capabilities designed to recover or restart workloads after certain host failures, usually using clustered resources.
12. Live migration
Moving a running VM to another compatible host with minimal or no interruption, subject to platform requirements.
13. Cluster
A group of connected hosts managed together to provide shared capacity and, in some designs, resilience.
14. Resource pool
A logical allocation of computing resources used to organize and control workloads.
15. Hyperconverged infrastructure (HCI)
An integrated approach combining virtualized compute with software-defined storage and infrastructure management.
16. Disaster recovery (DR)
Plans and technology for restoring applications and data after an outage or major disruption.
How to start evaluating server virtualization
You don’t need to begin with a complex migration. Start by understanding what your organization already runs and what success would look like.
- List your workloads. Document applications, operating systems, dependencies and current resource usage.
- Set non-negotiable requirements. Establish availability, security, performance, backup and support needs.
- Check compatibility. Validate guest OS support, hardware, backup tools and application vendor requirements.
- Compare complete costs. Include licensing, hardware, storage, training, migration and ongoing administration.
- Test before committing. Run a proof of concept with representative workloads and recovery scenarios.
Frequently asked questions
Is server virtualization the same as cloud computing?
No. Virtualization is a foundational technology that makes hardware resources available as VMs. Cloud computing adds service delivery, automation, on-demand provisioning and other operational capabilities.
Does virtualization make applications slower?
It introduces some overhead, but modern hypervisors can deliver strong performance. The real experience depends on hardware, storage, configuration and how much CPU and memory other VMs consume.
How many VMs can one physical server run?
There is no universal number. Capacity depends on available cores, memory, storage performance, workload demand, resilience headroom and product limits. Sizing should use measured workloads, not just VM counts.
Is server virtualization secure?
VM isolation helps separate workloads, but it does not eliminate risk. Hypervisor patching, access controls, network segmentation, monitoring and tested recovery procedures remain essential.
Sources & further reading
For deeper technical background and product-specific details, consult the following documentation:
