I'm back from Cloud Connect, a mid-sized cloud computing show held this week in Santa Clara. This was really a vendor and consultant conference, but there were a good number of end users there, and those actually deploying cloud-based systems which are few and far between these days.
On of the things that I noticed was lack of the S-Word, or SOA. While everyone agrees that you need to leverage SOA, in my discussion with those attending the presentations, the links from cloud computing to SOA were not clear at all. In other words, people are thinking about cloud computing without architectural context, which I think is a bit dangerous.
The core issue is that cloud computing is really a destination for an architecture, but it's not an architecture unto itself. The process of getting to a final architecture is about understanding existing systems, core business objectives, and then mapping a path forward using the best technology, including cloud computing.
The trouble with architecture is that it's almost as popular as the kid in the classroom that reminds the teacher to assign homework. However, if done properly SOA, when used with cloud computing, actually drives innovation, not stunts it.
The larger fear here is that those leveraging cloud computing without good architectural context are destined to find that cloud computing actually complicates their existing IT issues. This is really about a carefully planned configuration of IT resources that bring the best value to the business, not about what's hyped and what's popular. That's how many enterprises got into an architectural hole. It's time to stop digging.