Back to the future with VIRTUALISATION

There is momentum in the demand for virtualisation technologies, but don’t forget to invest in training to protect return on investment, writes GORDON SMITH

“You’re almost getting back to a mainframe operating environment – you can test in isolation, scale more easily but without the huge proprietary cost of running a mainframe,” explains Kevin Reid, business development manager with SureSkills, an IT advice and training provider. “With mainframes, it was the cost and complexity of running it that drove people away. With virtualisation, you’re talking about commodity hardware providing the capacity to perform and solutions from Microsoft, tems vendors or Sun; you’re not locked into a single operating tem or hardware vendor. You have the ability to run multiple tems in parallel, run Linux or Windows or Solaris all in the same hardware.” Although virtualisation itself is not new – SureSkills has been working with the technology for seven years – Reid says there is a momentum building in the market. “There’s a push to reduce the total cost of ownership of IT tems and one of those solutions is virtualisation.” Practising what it preaches, SureSkills has already converted 90pc of its own IT infrastructure to being virtualised. “We have built in disaster recovery and high availability, flexibility and scale. It gives so much flexibility to the organisation,” says Reid. Virtualisation also allows extra computing resources to be allocated quickly and easily to cope with increased demand.

It’s much more than technology for its own sake – it delivers a real business need, allowing organisations to be more agile and respond quickly to changes in the market. The old way would take a three- to six-month bedding down period to procure and install new hardware and test the application before it could go live. “But with a virtualised platform, provisioning the tem becomes a matter of minutes as it’s template-based,” Reid points out. Other benefits to the business centre on disaster recovery and high availability. Virtualised tems are easier to back up and to replicate in another site, so that companies can go on working in the event that one of the tems stops running. Clustering – spreading the computing load across multiple tems to avoid a single point of failure – used to be a costly exercise but virtualisation allows this without the expensive price tag, says Reid. “If you have a virtualised platform with more than one server, you have clustering by default because if one piece of hardware fails, the other one will take over the running. Testing new tems is also easy as a part of the virtualised environment can be set aside for this.”