What about on-premises computing?

We have been mesmerised by the large migration that’s been taking place as organisations move their workloads from on-premises deployments to off-premises. What happens to what remains behind?

Over the last few years the IT industry has been obsessed with the concept of cloud computing. In fact I can’t recall a time in recent years that I’ve browsed an IT news website and not seen at least one article about cloud. The amount of coverage may indeed be warranted but I can’t help but feel we have become somewhat distracted. There is still a future for on-premises computing.

In December, IDC Australia released our predictions for 2013 and one has received quite a lot of interest from both supply and demand sides of the market. The headline for the prediction is simple enough: On-Premises Computing will Catch a Second Wind with Converged Systems.

What I am saying in the prediction is that we have been mesmerised by the large migration that’s been taking place as organisations move their workloads from on-premises deployments to off-premises. What I am interested in is what remains behind.

With so much scrutiny placed on the workloads - and therefore infrastructure that remains - how are CIOs deploying in-house workloads? There are four main reasons for keeping your workloads in-house: innovation, a lower deployment cost, velocity and high business value.

In many cases we see a lower deployment cost for running workloads in-house rather than off-premises. However, even in this scenario the decision to keep a workload in-house is not straightforward as CFOs are watching IT spending very closely to ensure that all efforts are being undertaken to move away from capital expenditure. The result is that business cases attached to in-house technology deployments in 2013 will require far more rigour than before.

High business value is another driver for keeping workloads in-house. These are the workloads that are central to the organisation and may require uninterrupted delivery of compute or are custom applications that have been developed over a period of years - often decades - and require resourcing and management that is best kept in-house.

I’ve been to many vendor events over the past few years espousing the value of private clouds but here’s something the vendors won’t tell you: private clouds are extremely difficult, expensive and time consuming to build. Even with the right skills and large budgets, the time required to deploy private clouds can rule them out from selection criteria. Converged systems offer a way of short-circuiting these long lead times.

The last of the major drivers is innovation, which is something I liken to cholesterol in an IT department. Cholesterol is essential to our good health and wellbeing but too much leads to serious issues. In our IT departments, getting the right level and balance of innovation is a priority as every time we innovate we amplify complexity, which leads to increased expense.

Converged systems enable organisations to direct innovation appropriately ie, to where innovation will have the greatest impact; the software and services that tie together the various layers and connect to the organisation. A change in the Australian server market has been long overdue. Research tells us that we are still a long way from achieving management and efficiency standards required to drive costs down in our organisations. The good news is that with Australian organisations being one of the world’s fastest to take up new technologies, we’ll be watching closely, as in 2013, 1/3 of all server deployments in Australia will be converged systems.

*Matt Oostveen is the head of research for IDC Australia overseeing all research divisions. Based in Sydney he has more than 15 years of information technology experience in the Asia-Pacific region. He is responsible for identifying emerging trends, new business and pricing models, and assessing the interconnecting relationships between hardware and software technologies from a services solution perspective. He also runs end-user research into demands and adoption trends of CIOs across Australia.