There has been an increasing frenzy of media coverage regarding Cloud Computing recently. It’s a phrase that is thrown around for everything from IaaS (Infrastructure as a Service), PaaS (Platform as a Service), SaaS (Software as a Service) and just about everything in between.
So what IS going to happen? The real answer is, I’m not sure and I’m certain that nobody else does, but I know what makes sense with the technologies that are now available and as a response to the pressures of the current economic climate.
In the past, companies have been found comfort in designing, purchasing and implementing their own technology infrastructure. There has been a certain fuzzy feeling in being able to touch and feel your own hardware, to press a button to power cycle a hung server and to be able to go and look at the lights flashing away on the front panel. IT Managers could close the door, secure in the knowledge that the magnetic lock on the Comms Room door would ‘click’ and keep everything snug and safe inside. This was a necessary scenario in the past, but it created huge challenges for technology managers. First there needed to be a justification of large budgets to implement the initial installation as well as the need to create an environment that kept things cool and full of juice, then came the annual budget justification to refresh the technology to meet the increasing demands of the business, which was generally met with the statement “but we only just bought new technology?” by members of senior management. I won’t even start on the never ending fight to keep servers cool and provisioned with enough power while you jump through the building regulation hoops to locate condensers on roofs and to route extra power and data through an already crammed riser… It became a never ending battle to keep things up and running and generally kept the focus on keeping the lights on and away from growing the business.
Surely there has to be a better way… Well, with the reduction in the price of high-bandwidth communications lines, the maturing of the processes and policies used by hosting companies and the increasing functionality and stability of virtualisation technologies, there is. It takes a change of mindset, but if technology leaders can disconnect themselves from the need to physically ‘connect’ with their infrastructure, they then can free themselves from the shackles of being a Jack of All Trades so they can concentrate on building services that are focused on driving the business forward.
So how do you achieve this? There are a number of companies emerging in the market who have evolved from being purely DataCenters, hosting physical machines and providing hands on support, to providing Infrastructure-as-a-Service, that uses Virtualisation as the core of the services, but with the ability to incorporate physical hardware in to the infrastructure providing companies with the option of tailoring around the edges for their specific business needs. Let’s be honest, we all do the same thing for the everyday services we provide to our users. We provide email and storage, we make sure that data is backed up and available in the event of a disaster, we work hard to make sure that PCs run quickly and run the core applications to allow users to create documents, spreadsheets, presentations and to be able to communicate through an email or chat client with the big wide world. We spend a lot of time and effort, employing the right people to make sure it all works together. We take hours making sure that we use the right hardware and we scratch our heads making sure it’s all safe and secure. But isn’t everyone doing this? Yes they are which is why a hosted, managed solution makes sense. A good hosted service will provide the core functions necessary to run a network, in modules that you can pick and choose to build your infrastructure, which is then charged on a usage basis, much like you do for your phone or electricity bill. You wouldn’t dream of building a private phone system and put extensions in the locations of those you want to call and you wouldn’t build your own personal power station either so why do you need to replicate the standard functions in your infrastructure?
So what are the benefits? First, you remove the ongoing upkeep of a growing infrastructure. No more one-off expenditure, no more support contract renewals, no more managing the resource necessary to make sure that specialist hardware is running optimally. All of the ongoing costs are built in to your scheduled payments, but with the benefit of the provider’s buying power due to the economies of scale. Second, you no longer have to worry about environmental issues. Your slice of infrastructure is sitting in purpose built DataCenters so it’s always going to be at the right temperature with the right level of power. Third, your infrastructure is now flexible. If you grow you add, if you shrink, you remove and you only ever pay for ‘what you eat’. You no longer have to worry about having expensive hardware sitting dormant due to a failed project or venture. Fourth, you don’t have to worry about human resourcing. The day-to-day management takes place as part of the provider’s processes and as a result of the SLAs in place, when you have a request at 2am on a Sunday morning, there’ll be somebody at the end of a phone willing to help. No more fretting about waking somebody up. Finally, although you have to worry about Disaster Recovery and Business Continuity Planning, there will already be redundancy built in to your solution. It’s worth checking, but most reputable hosting providers will mirror their infrastructure to ensure that they meet SLAs. Remote access with varying levels of security will also be provided, so if your office is out of bounds, your employees will have the ability to access systems from their home locations.
“Hold on”, I hear you ask, “Doesn’t this mean that I need to install huge bandwidth between my offices and the hosting provider?” The answer is probably not. Your processing is all localised in the hosting provider’s DataCenter, so the only traffic that flows between them and your offices is keystrokes, video and small quantities of miscellaneous data. The reality is that for a 20 person office, 10Mb of bandwidth, with some sort of low grade symmetric-comms line as backup, is likely to be more than enough.
This all sounds too good to be true doesn’t it? Yes it does, but this sort of scenario is achievable with services that are available now. I’ve used some very broad brushstrokes to paint the picture above, but the theory is legitimate. It will need plenty of thought and bundles of planning, but it is a realistic goal for your technology infrastructure. There are specialists who can help you understand the benefits that can be achieved from moving your technology in this direction and who will help you to migrate from your current situation without disruption to your users and to ultimately reach your goals.
The first step is to make some enquiries. Have a look at what is available and investigate the offerings on the market. Don’t rush in to the first provider that you come across, make sure that they have technology that is similar or better to that which you would install yourself. Plan, plan and then plan some more and some day the technology nirvana described in this report may one day be yours.