I've been hearing about some disturbing things out there. The fact is, many companies are creating separate architecture teams for the new push into the cloud computing and those teams are not working with the existing SOA teams. The rationale I'm hearing is that they consider cloud computing to be "new," and thus believe it needs a new team, with its own budget and different leadership.
However, those of you who think you can separate architecture and cloud computing are gravely mistaken. Indeed, in all of the projects that I'm working these days the best way to drive toward cloud computing is really to leverage SOA approaches. This means understanding your problem domain at the data, services, and process levels before moving data, services, and processes out to cloud computing platforms.
[ Go beyond the hype. Learn what cloud computing really means. | IT groups that understand SOA may be able to take better advantage of the cloud. | Follow the cloud with whurley's Cloud Computing blog. ]
The end-state architecture is going to be a mix of on-premise and cloud computing platforms, and thus the architecture will span on-premise and cloud computing systems. Let's get a clue: It's architecture, not cloud computing, that will save the day here.
I suspect this kind of stuff will continue to pop up as those in IT, in some cases, have yet to gain perspective in light of the hype around cloud computing. Don't get me wrong, cloud computing is a great way to go if there is an architectural fit. Indeed, I'm writing a book on that topic right now. However, like any technology approach, you have to consider it an "architectural option" and not "the architecture."
Trouble comes in when the cloud computing team lives up to its namesake and just focuses on creating new processes and moving existing processes to cloud computing. The fact of the matter is that really solves nothing unless there is complete synergy with the existing enterprise systems, and in too many cases that seems to be an afterthought.