Types of Server Virtualization
Types of Server Virtualization
Full-Virtualization
A hypervisor is a specialized type of software that is necessary for full server
virtualization. There are many different hypervisors available, so make sure to
compare your requirements with available features before finalizing a purchase.
Hypervisors function by communicating with servers to monitor disk space and CPU
usage. The software can then allocate resources as needed across multiple virtual
server deployments. Computing power can even be shared across various
operating systems with ease without sacrificing efficiency. Hypervisors do require a
set amount of dedicated server resources to operate, limiting overall server power.
When utilizing full virtualization, the separate server instances will not need to be
aware of each other.
Para-Virtualization
The major difference between full and para-virtualization is that each OS on the
server is aware of each other's presence. This allows for an entire network to work
together to manage resources. The most significant benefit of para-virtualization is
that the hypervisor doesn't require nearly as many resources to operate because
different virtual servers are aware of each other and thus can more efficiently share
physical resources.
OS-Level Virtualization
The most basic form of server virtualization is OS-level virtualization. There are
limitations when using this method, but it can be implemented and maintained with
fewer resources than para- or full- virtualization. When using OS-level virtualization,
there is no need for a hypervisor. The duty of managing resources and separating
virtual machines is instead handled by the physical server’s operating system. The
drawback to this solution is that each virtual machine will have to run the same
operating system, because the OS is acting as a hypervisor.