I think decision makers are pushing stuff towards the cloud because the complexity and management overhead of internal infrastructure is overwhelming. There’s this view that “put it in the cloud” makes all the pain go away, but it doesn’t. Someone still has to buy storage, bandwidth, CPU, RAM, redundant infrastructure, a continuity plan. How well thought-through are the continuity plans for IaaS? Some are very well done, I know, but often only at the cost of reintroducing the very complexity people were trying to manage out in the first place. It’s taken me a week not to get a fairly simple pricing request turned round by RackSpace because, I think, of the questions around continuity. Not that I’m singling them out, I think everyone has the same problems (I know our own internal cloud sales team does).
Of course once you realise that the whole aim is to abstract your company from the messy business of providing infrastructure, the idea of cloud and IaaS starts to look a lot like something we have seen before: outsourcing. I wonder if perhaps cloud is, in part at least, simply the most successful rebranding exercise in history?