Cloud Computing is a very popular phrase these days, and it's been creeping its way into SQL Server circles with the introduction of SQL Server Data Services and other service offerings from Microsoft and other companies. It seems to be what many IT managers and some IT professionals want to talk about.
But it's always struck me as being a little odd. After all, I've never met either:
A) Software that runs on water vapor and ice particles suspended in air,
B) A cloud that's made up of high-end computer hardware.
Yes, I know - that's not really fair. Everyone knows that the "cloud" in Cloud Computing refers to the Internet, which is usually drawn as a big fluffy cloud on whiteboards because it's too complex and dynamic in structure to be diagrammed accurately at more detailed levels of abstraction. But still, it seems to me that too many people fail to look deeper, to force the "cloud" abstraction to leak, as it were.
Until it leaks on them.
Obviously, having your software run "in the cloud" means that it's running in a well-connected data center, probably one that's managed by a 3rd party. But the cloud is still made up of hardware and software, and regardless of who owns and operates it, it still needs to be managed by someone. And what happens when that fails?
Well, bad things, right? I came across this article on The Register today:
What's the moral of the story? I'm not going to claim that I know, but I'm sure that it is something related to "good for the goose, good for the gander" or the like. The same bad things that can happen to your own data center can happen to others' data centers as well, even when your applications and services are running there.
So is it time to look at your SLA again?
 My favorite source for bitter and sarcastic IT news
 Ouch and a half.