Case study: Virtualisation at the London Borough of Hillingdon

Our municipal buildings date back to the 1970s, and use power supplies from the same period. Since moving in there we had simply gone out and bought the...

Our municipal buildings date back to the 1970s, and use power supplies from the same period. Since moving in there we had simply gone out and bought the hardware we needed, when we needed it. The result was that we actually had no idea how many servers we had on site or how full they were. However, one thing was for certain; our building was completely out of power and, if we kept adding hardware in the way we had done, roads would literally have to be dug up and new power cables put in to keep up with demand, writes Roger Bearpark, assistant head of ICT at the London Borough of Hillingdon.

Unfortunately, we faced 100% year-on-year data growth, driven mainly by the growth of employees' email boxes, the need to retain documents such as benefits assessments records, and the increased use of digital images in planning applications and property and highway inspections - so another increase in hardware seemed on the cards.

I'd read a fair amount about server and storage virtualisation in the past, but it had always seemed a fringe technology and, from a public sector perspective, a potentially conservative and disruptive choice. Some of my peers had begun talking about using virtualisation in test and development environments, but we decided to take the plunge and launch it into our live environment. It was a decision that filled my team with equal measures of excitement and trepidation.

Within six months of putting out our first tender, we had the first of our VMware virtual server environment in operation, closely followed by two virtualised Compellent SANs. I have to be honest and say that from the outset, nobody was quite sure what the outcome was going to be, but we structured the roll-out team in such a way that everyone got to experience and learn about each part of the installation programme - helping to eliminate any potential fears and letting all the team take real ownership of managing the new server and storage platforms.

Looking back, the experience has been very much like stepping into a new car; you know there's a powerful engine underneath, and you know you can put your foot to the floor and go faster than you've done before. However, it takes some time to get used to the controls and get the confidence to see what it can really do. Metaphorically speaking, when we put our foot down after all our familiarisation, we weren't disappointed at all.

As a validation of my own faith and that of my colleagues, we've won and been finalists a number of IT industry awards over the past six months. I shouldn't brag, but it's not all that surprising given that we've saved around £20,000 alone by reducing power consumption from 34 kWh to 1.1 kWh. We also reduced server hardware by 97% and reduced disk space requirements by 45%, saving a further £6,500 a year in power costs by moving data onto less expensive, energy-efficient drives - all of which also helps us do our bit for the environment. In addition, we saved a further £50,000 as we didn't have to expand the IT team to deal with our growing data volumes - good news for these economically uncertain times.

Using the host of tools inherent within the Compellent solution we have worked to establish a very cost effective and technically advanced Business Continuity solution that also offers our PCT neighbours a platform upon which to build their own continuity plans. This provides disaster recovery functionality at a fraction of the cost of other replication solutions and based on a technology we understand and trust.

As you might have gathered, I get very frustrated when IT professionals are reluctant to take risks, and I think here we have shown that a little bit of bravery and courage to try something new can go a long way.

Read more on Server hardware