Virtualisation refers to technologies designed to provide a layer of abstraction between computer hardware systems and the software running on them. By providing a logical view of computing resources, rather than a physical view, virtualisation solutions make it possible to do a couple of very useful things: They can allow you, essentially, to trick your operating systems into thinking that a group of servers is a single pool of computing resources. And they can allow you to run multiple operating systems simultaneously on a single machine.
Beyond the potentially dramatic cost savings, virtualisation can greatly enhance an organisation's business agility. Companies that employ clustering, partitioning, workload management and other virtualisation techniques to configure groups of servers into reusable pools of resources are better-positioned to respond to the changing demands their businesses place on those resources.
Also, this technology offers the potential for a fundamental change in the way IT leaders think about computing resources. When managing individual boxes becomes less of a challenge, the focus of IT can shift from the technology to the services the technology can provide.
Cost benefits
IT departments everywhere are being asked to do more with less, and the name of the game today is resource utilisation. Virtualisation technologies offer a direct and readily quantifiable means of achieving that mandate by collecting disparate computing resources into pools that can be shared.
For example, analysts estimate that the average enterprise utilises somewhere between five per cent and 25 per cent of its server capacity. In those companies, most of the power consumed by their hardware is just heating the room in idle cycles. Employing virtualisation technology to consolidate under-utilised x86 servers in the data centre yields both an immediate, one-time cost saving and potentially significant ongoing savings.
The most obvious immediate impact here comes from a reduction in the number of servers in the data centre. Fewer machines means less daily power consumption, both from the servers themselves and the cooling systems that companies must operate and maintain to keep them from overheating.
Turning a swarm of servers into a seamless computing pool can also lessen the scope of future hardware expenditures, while putting the economies of things like utility pricing models and pay-per-use plans on the table. Moreover, a server virtualisation strategy can open up valuable rack space, giving a company room to grow.
From a human resources standpoint, a sleeker server farm makes it possible to improve the deployment of administrators.
What to look for
What should you look for in a virtualisation solution? In a word: management. The core hypervisor technology that decouples the application stack from the underlying hardware is well on its way to commoditisation. Look for solutions that provide easy-to-use tools for gathering statistics and applying dynamic policies to better allocate your physical resources among the virtual consumers of those resources.
Virtualisation can go a long way towards reducing the physical requirements of the data centre, but it can also compound the level of management complexity of those servers. So look for solutions that provide cross-platform systems management for both the virtual and physical machines.
Also, you'll want the ability to migrate your organisation's legacy applications and existing operating systems, without modification, onto virtual partitions. This migration should make it simpler to enhance the performance of those applications, but you'll need a solution that supports the integration of virtualisation with legacy management tools.
