Traditionally in an average SME, the term ‘server’ refers to a big noisy computer, a lump of tin in a room somewhere out of sight within the business premises. The more servers a company has, the more lumps of tin they have sitting in the room, each pulling electricity and generating heat.
The concept of server virtualisation began in the 1960s, as a method of dividing resources between different applications within a mainframe. But as with most technologies, SMEs generally didn’t begin making use of the idea until a few decades later.
Virtualisation allows multiple servers to run from one physical computer, known as a ‘host’. The host is a powerful machine with plenty of resources (processor power, random access memory and storage capacity). Installed on the host is a ‘hypervisor’, a piece of software that allows for the simulation of computers. These simulated computers are known as VMs (virtual machines). Each VM is allocated a portion of the available resources on the host. If we install a server operating system onto a VM, we have a ‘virtual server’. We can therefore have multiple servers, each performing different tasks, all running from the same physical host computer. Not only does this save space, it also reduces costs in terms of energy consumption, but there are other advantages.
A virtualised server can be remotely managed from the hypervisor, as if the person managing it were in the server room looking at a physical server. That is, if a server becomes unresponsive, it can be forcibly powered off from the hypervisor. Memory and other resources can be added and removed from a server without the engineer performing the task needing to be anywhere near the physical equipment. Management of servers suddenly becomes a lot easier.
There are a number of vendors on the market providing hypervisor software. The most successful of these in the early days was VMware. In recent years however, Microsoft have made significant progress with their own offering; their ‘Hyper-V’ product is now a force to be reckoned with and provides a very competitive price point compared to its rivals.
Among a range of many useful features, Hyper-V provides the option of ‘replication’, where VMs are replicated at fixed intervals (30 seconds, 5 minutes or 15 minutes) between two hosts. This is an excellent tool in the arsenal of any SME’s business continuity plan. It means that in the event of a critical hardware failure on the primary host, VMs can be ‘failed over’ to the replica host and a business can be back up and running within minutes, rather than hours. The replica host can also be placed in a different physical location to the primary host, providing protection against data loss through fire or flooding. As of Microsoft’s Windows Server 2012 R2, ‘extended replication’ is also available, giving SMEs the option of replicating to a tertiary host, at yet another physical location.
SMEs can quickly take advantage of the benefits of virtualisation by performing a ‘physical to virtual’ (P2V) conversion of their existing physical servers. This procedure is usually carried out during an agreed period of downtime (on an evening or weekend, for example). Within a few hours, an SMEs whole server infrastructure can be up and running under the management of a hypervisor, with all the benefits that this brings.
Trenton Business Systems