In the '60s, all the windows in the data centre were made of bulletproof glass and the guards waiting at the main entrance did, in fact, carry firearms. You'd walk through about 10 doors and a mile of corridors and then enter the sacred home of the System 360, which was IBM's largest and fastest mainframe computer. You could barely speak above the roar of the cooling units, whirring of the disk drives, and spinning of the tape reels. Overkill? Perhaps, but one thing you could count on, was that this computer would never stop running and was built to last. Then came the PC frenzy of the `80s and `90s, where PCs ruled the roost and corporations began replacing their mainframes with blade servers the sizes of pizza boxes. The purchases were easy to justify; not only were the units physically smaller, but they generated less power and used less heat. Life seemed to have gotten simpler. Cheap computing became the order of the day. Something, however, went wrong and computing began to run amok. Computer crashes and downtime became the norm, not to mention viruses that could make an entire system crumble. Rebellious end-users had taken ownership of the computing function and IT departments were left fuming. Fast forward to 2009 and far from the Paleolithic systems that have been lost to the evolutionary shuffle, data centres are chugging away and due for a comeback. IT and business get along much better, although economic times tend to encourage that. IT departments bite their collective tongues and declare: "Nice of you to welcome us back, but this time we are going to do it our way." In thousands of organisations around the world this scenario has played itself out. The truth be told: the data centre, and its principal resident, the mainframe computer, never went away, it just slipped into the background. Now it's time for a comeback. Indeed, new global research reveals that 84 percent of technology organisations are planning to implement a data centre transformation (DCT) project within the next year, primarily to lower costs, but also to reduce business risk, according to a Hewlett-Packard sponsored report called "Data Centre Transformation: Key Implementation Drivers". The new research indicates that 20 percent of technology decision makers are initiating a complete transformation, while the remaining 80 percent are implementing individual transformation projects. In other words, data centres and their mainframe residents are alive and kicking. Senior executives (CIOs) are taking dead aim at the data centre to control costs and quickly achieve returns on technology investments, citing security and the need to backup operations. IT has come full circle. "We have seen a renewed interest in data centre technology from both a cost and asset point of view," says John Bennett, worldwide director, Data Centre Transformation Solutions, HP. "In 2009, the realisation that you can transform your data centre for strategic advantage and reduce your costs is a very intriguing value proposition. You can have your cake and eat it too." The number of data centre revitalisation projects around the world numbers in the thousands. Hewlett-Packard itself went from 85 data centres to 6, reducing operational costs by 50 percent which translates into millions of dollars saved, and if for that reason alone, senior executives are paying attention. Energy efficiency, and data centre consolidation are usually at the core for such efforts, "power and cooling tend to be targeted for large savings," Servers have a major role to play, but so do mainframes. According to Bennett, most large organisations have found a nice balance of the old and the new, and reason is simple. In a 7 X 24 world, no company, large or small, can afford downtime. Typical of this story is Ash City, a Toronto, Ontario-area online clothing retailer, which recently moved its entire data centre lock, stock and barrel to a location further outside the city. Ash City's fully functional data centre was built in less than a week; something that normally takes months to construct. It houses the company's servers in high-density server racks that are cooled with an efficient chilled water system. During cooler months, which are not uncommon in Canada, a heat exchanger provides green, low-cost cooling by using the external cold air to provide cooling. "The number one challenge was keeping our business up and running and always online," says Michael Suen, CIO, Ash City. The relocation of the servers and equipment occurred over a number of weekends, allowing Ash City to continue business without interruption. In the industry, conversion projects are never easy, and can be likened to doing a heart transplant while running a marathon, or doing an engine replacement while in a Formula One race. Part of the difficulty, says Osama Arafat, president of Q9 Networks, an international provider of third-party data centre services: "Most data centres need to operate 7 X 24, most companies cannot afford to have that much staff. When something happens with the HVAC system or there is a security breach at 3:00 in the morning, somebody has to react to it." Another factor is the push to green computing, says Arafat. That data centres are much smaller and therefore cost a lot less is somewhat of a fallacy. "Yes, the real estate cost-savings associated with a smaller footprint are there. The power used has gone up exponentially. So while it is smaller from a physical perspective, the overwhelming cost has to do with the energy factor." According to Arafat, power and HVAC costs are what really matters and that has opened the executive's eyes to what really matters. "Every single day, organisations are getting more dependent on their computing infrastructures. The more expertise is required, the more monitoring and the more response, and the greater the impact on the organisation," he says. "In the last year or two, I've seen a very interesting shift especially at the CEO level, from seeing IT as a black-hole that wants more money, to maybe that's where I should look for my competitive advantage." Darin Stahl, lead analyst, Info-Tech Research in London, Ontario says that "there is a certain amount of hype around data centre transformation, but what it really is about is bang for the buck." And it's not just the bigger companies. "It's not an exaggeration to say that a midsise company of $10 to $20 million could easily spend $1 million on IT. For those companies, it's all about backing up their IT operations and using disaster recovery solutions to get back up and running quickly. "Mainframes don't rule the world like the dinosaurs once did but they are still around," says Stahl. Computers have gotten smaller and manufactures have learned to adapt their designs, and for them, a modular data centre or data centre in a box have become viable solutions. "You can get a baby mainframe for a $150,000." For all the renewed interests in data centres, Stahl says the "trigger points" are reduced operating costs, disaster recovery and the need to stay online all the time. "There's a lot of heat issues involved, there's a lot of power issues involved. Now it's way more complicated. You just can't fill a room with cool air when you start moving servers around. You have to look at precision cooling systems and metered systems, thermostats. There is a lot of new complexity here." The problem? "Those skills around mainframe technology and building an IT architecture may have atrophied and organisations may need to go back and revisit some of those lost skill sets," says Stahl. The old data centre is back, and for the traditional IT department, the news could not be better.