- Server virtualisation enables multiple virtual servers to run on a single physical server, maximising resource utilisation and reducing costs.
- The main types of server virtualisation include full virtualisation, para-virtualisation, and OS-level virtualisation.
What is server virtualisation?
Server virtualisation is a technology that allows a single physical server to host multiple virtual servers, each functioning as if it were a standalone physical server. This process is achieved by using software called a hypervisor, which divides the physical server’s resources — such as CPU, memory, and storage — into separate, isolated environments known as virtual machines (VMs). Each VM can run its own operating system and applications, independent of the others.
1. Hypervisor role: The hypervisor acts as a bridge between the physical hardware and the virtual servers, managing resource allocation and ensuring that each VM operates efficiently without interfering with others. There are two types of hypervisors: Type 1, or bare-metal hypervisors, which run directly on the hardware. Type 2, or hosted hypervisors, which run on top of a host operating system.
2. Resource optimisation: By creating multiple VMs on a single physical server, organisations can optimise their hardware usage. Instead of having several underutilised servers, they can maximise the capacity of a single server, leading to better resource utilisation and reduced hardware costs.
Also read: How to build an internet server?
Also read: Kaldera: Pioneering digital solutions in the Indian Ocean
Types of server virtualisation
There are three main types of server virtualisation: full virtualisation, para-virtualisation, and OS-level virtualisation. The primary difference between these types lies in how isolated each virtual machine (VM) is from the others, and how they interact with the host server.
1. Full virtualisation: This type of server virtualisation creates multiple VMs that are entirely isolated from each other. Each VM acts independently, with no awareness of the other VMs on the host server. The hypervisor, known as a bare-metal hypervisor, is installed directly on the physical hardware. It acts as a layer between the hardware and the VMs, each with its unique operating system. This setup allows for high isolation and security but can sometimes lead to slower performance due to the hypervisor’s resource requirements.
2. Para-virtualisation: Unlike full virtualisation, para-virtualisation allows VMs to be somewhat aware of each other and work together across the network. This type of virtualisation reduces the workload on the hypervisor, as the VMs themselves help manage resource allocation. Para-virtualisation requires the host server to have an operating system that collaborates with the hypervisor through hypercall commands. This setup was developed to address the performance issues of early bare-metal hypervisors, but it is less common today because modern servers have improved support for full virtualisation.
3. OS-level virtualisation: Also known as hosted virtualisation, this type eliminates the need for a traditional hypervisor. Instead, the host server’s operating system handles the virtualisation tasks. All VMs must use the same operating system as the host, sharing its common binaries and libraries. OS-level virtualisation can support a large number of VMs on a single server with fewer resources, making it an efficient option. However, it has a significant drawback: it creates a single point of failure. If the host OS fails or is compromised, all VMs are affected, potentially leading to widespread downtime.
