A Guide to Key Virtualization Terminology — Part 2

One of the most important steps in mastering a new technology is learning the associated terminology or vocabulary. In the IT field, this can be a very challenging step, as much of the terminology is often used inconsistently. Here we will define the terms associated with IT virtualization. It is mainly vendor-agnostic, but it does provide some vendor-specific terms, product names, and feature names used by VMware, Citrix, and Microsoft.

1. Clone – Typically refers to the action of copying one VM or VM template to create a new VM. During a clone operation, the VM files are typically copied, renamed, and modified to customize the new VM.
2. VM Snapshot – A point-in-time capture of the state of a VM. Snapshots allow the user to revert the VM to a previously captured state. A primary use is to undo changes that were made in a VM but are no longer wanted.
3. VM Migration – The movement of VMs from one resource to another, such as from host to host or datastore to datastore.
4. Live VM Migration – Occurs while the VM is running.
5. Cold VM Migration – Occurs while the VM is shut down.
6. Resource Contention – A state where a VM is competing for a scarce resource with other VMs or overhead processes. For example, if the memory capacity of a host is currently fully utilized, and some VMs attempt to demand more memory, then memory contention occurs and some VMs may begin to swap to disk.
7. Highly Available (HA) – A system or component that has some automatic protection in case of disruption. The protection may allow a small amount of unplanned downtime, but it will automatically correct the issue within a pre-determined time interval.
8. VM High Availability (VM HA) – Ensures that a VM is automatically made available, although the host on which it runs fails. VM HA may require an automatic reboot of the VM on another host.
9. Fault-Tolerant – A system or component that has automatic, state-full protection in case of failure. For example, some software applications are designed to replicate state to multiple servers and databases to provide a fault-tolerant application. The failure of a server does not result in any loss of state of the application or any disruption to the end user.
10. Fault-Tolerant VM – A VM that continues to run state-fully even if host hardware fails. This may be achieved by synchronizing the execution and state of multiple VMs running on multiple hosts.
11. Provisioned Over-Commitment – A measurement, usually in percentages, by which the amount of provisioned virtual hardware is greater than the actual physical resources. For example, if a set of thin provisioned virtual disks is configured for a total of 3 TB, but the datastore where they reside is only 2 TB, then the over-commitment is 150 percent.
12. Over-Committed –A state where the actual, attempted resource usage exceeds the capacity of the actual hardware resources. For example, if a set of VMs stored in the same datastore generate more I/O than the underlying LUN can accommodate, then the datastore is over-committed.
13. Local VMs – VMs executed directly on a client system, such as the user’s PC. Some virtualization products, such as VMware Workstation and Microsoft Virtual PC, are designed solely for running local VMs. Some VDI products allow virtual desktops to run remotely in the datacenter, but also allow the user to check out a virtual desktop, execute the virtual desktop locally on a client system, and check in the virtual desktop upon completion.
14. Thin Client – A client device that has a very lean implementation of Windows or Linux and is mainly intended to allow the user to connect to a remote virtual desktop rather than to run applications natively.
15. Zero Client – A client device that is even leaner than a thin client. Typically, a zero client runs an embedded, proprietary OS and has no local disk. It is used to connect to remote virtual desktops.
16. Virtual Machine Manager (VMM, also called a virtual machine monitor) – A process that controls the execution of a VM and brokers its use of virtual hardware with the underlying host. It notifies the host when the VM needs to access the physical resources.
17. CPU Hardware-Assisted Virtualization – Features that are commonly provided on modern CPUs, allowing the host to offload some of the virtualization work to the CPU to improve performance.
18. Virtualized Application – A packaged software application that runs in a virtualized, runtime environment, where the application perceives that it is natively installed. For example, a virtualized Windows-based application accesses a virtual Windows registry and virtual file system that are created at runtime by the runtime environment by overlaying modifications in the package on the native registry and file system.
19. Virtual Appliance – A pre-built VM containing pre-installed software that can be easily implemented. Typically, the appliance is downloaded from a website as an OVF file, deployed into the virtual infrastructure, and easily configured using the console of the VM and a web browser. Most virtual appliances allow very simple implementation, relieving the customer of a complex installation and configuration.
20. Open VM Format (OVF) – A specification that can be used to export and import VMs from one virtual environment to another. Typically, virtual appliances are stored in the OVF format.

Reproduced from Global Knowledge White Paper: vTerminology: A Guide to Key Virtualization Terminology.