My Way: Edmund Nuttall
Civil engineering firm Edmund Nuttall has taken advantage of a head office refit to virtualise its IT- and progress its Green commitment. The company's head of IT services Steve Shepherd tells the story to E&T.
E&T: How did your major server virtualisation project come about?
Steve Shepherd: Moving through 2007, I started to give consideration to trying to reduce our server footprint. At that time we had around about 50 physical servers. I had attended a number of seminars on virtualisation, primarily organised by VMWare, and that started the ball rolling as to whether we as an organisation could reduce our footprint of servers. I am very aware of our carbon footprint and wanted to see if there was any possibility of reducing it.
E&T: What was the next step in getting the project rolling?
SS: There were some internal discussions, I put forward various presentations and the idea went through a couple of committee meetings because, naturally, with this type of project, there is a cost. The idea was that from those 50 servers we could get down to a sizable number - at that point I had in my mind the idea of 10 physical servers. This went on and the seed of the idea started to grow in the strategy committee.
E&T: So what was next toward realising the idea?
SS: Then it was announced that the office in Camberley - which is on four floors - was to be completely refurbished. Part of the refurbishment is a complete rewiring of the office, and more fundamentally in my position, we've decided to move the server room from the second floor to the first floor.
That means all the telecommunications systems will have to be moved in stages. But it gave more emphasis to the importance of server virtualisation, as instead of having to move 50 physical servers, we're able to consider moving a lesser number.
E&T: So the timing was fortuitous?
SS: It was extremely useful timing. The other 'push' is that the new server room is probably one-third the size of the old room, meaning office space is recovered for use, so there's a benefit there.
E&T: When did the project get underway?
SS: The project was cleared to proceed last November, and then we tendered for a partner to assist us, as my team had a fairly limited knowledge of what was needed to be done at that point in time. We knew what was possible, but we didn't really have the technical skills to undertake that. We went through an elimination process and eventually decided on a partner to be used that was the best fit for Nuttall, and this was FoundationIT. My philosophy and policy is that whoever we work with externally, we treat as equal partners, and we try and have a very good working relationship.
E&T: When was the agreement signed?
SS: The agreement was signed in November 2007, and we ordered necessary Storage Area Network (SAN) technology in December. In January (2008) it was delivered. In the meantime we needed to go through a capacity planning exercise. This looks at our server environment: what each server is doing, what percentage is being utilised, and so on.
E&T: And how long did this take?
SS: A month, to look at the usage peaks and troughs. That ran through December to January.
During that time we discussed with FoundationIT how we should work together to implement this solution, and one of the conclusions was that it was essential we arrange training of our own personnel in VMWare. The way I arrange it, is that one person goes on a course and becomes the trainer for everybody else. This is my standard approach - and it saves money as well.
E&T: Talk us through how things progressed from January.
SS: The hardware had arrived, so through February - post-training - my team worked with FoundationIT to get a better understanding how the hardware and the software worked by setting up virtual machines, pulling plugs out to see what would happen, and so on; in other words, to test the system to destruction.
There were a lot of criteria to work through to make sure that everybody was happy, but that's so when we eventually get into a production environment, we've got a sound and secure system.
E&T: What about the switch-over between the old and the new systems?
SS: We had the advantage because we were still running production in our old environment.
When March hit we were ready to begin the server virtualisation process. By this time our servers had had to increase to 55 due to business needs, but through March and April we've brought the number of physical servers down to an impressive five, plus our SAN. So from start to finish it's been six months, but a lot of that time was spent on getting equipment, training. The actual server virtualisation only really took about two months.
E&T: That's pretty impressive. Do you think it helped that your staff 'trained' themselves?
SS: Very much so; I'm proud to say I've got very committed staff. They've been working out of hours because we still have to run production during the day and then migrate the existing physical servers into a virtual environment during the evening or weekends. The migration of a server can take a couple of hours or up to 24 hours - depending on what data needs to be transferred. There was a lot of commitment to working unsociable hours.
E&T: So you've achieved cost savings and shown commitment to being 'green' too with this project?
SS: Yes, definitely.
E&T: You sound very sure!
SS: Nuttall's is very, very green. Not just from a logo colour point of view, but as an organisation. I receive reports talking about corporate social responsibility and carbon footprints, so there is a big push to be aware of that over the next year; I'll be heavily monitoring our power usage.
E&T: Does this affect how you work - do you have a different kind of mindset on how you approach projects?
SS: I don't know about a different mindset; you are just aware of it and there is sufficient internal education on the subject. I think people do it automatically as they work. It's the overall general awareness to start off with, and then people view and come up with their own conclusions.
E&T: So what's next for you and your team?
SS: Our next challenge is improvement of our disaster recovery for ICT services. We've already identified how we're going to do that; we're going to replicate our server environment to an area office. We do have a disaster recovery policy in place, but we're looking to improve on that, so that's our next major project.
Curriculum vitae: Steve Shepherd
Worked in the ICT industry for nearly 40 years. Two main employers.
Early employment history: COBOL programmer for a city-based insurance company. After a couple of years was offered a job by Mears Construction to redevelop its systems from a GE computer to a Honeywell DPS4. Role changed to an Analyst/Programmer.
His redevelopment was a long-term project. Some years into the project Mears decided to relocate their business to Swindon. The development of systems continued within the company. Steve Shepherd become chief systems analyst with a team of five analyst/programmers reporting to him.
Mears applications: a fully integrated solution consisting of solutions for estimating, purchasing, personnel and payroll, a full financial management system and a job costing system.
When Mears went into receivership its assets were purchased by Edmund Nuttall (owned by HBG, a Dutch construction company).
Following a decision by the HBG IT director to merge the Nuttall IT department with that of Mears IT - a reverse takeover - Shepherd decided to stay with the new organisation as IT manager.
He now has a team of 20, split into Development, Administration and Support.