Cloud computing is much ado although not strictly about nothing. There are a good many instances where the bean counters will be only too happy to replace dealing with those pesky sysadmins with dealing with other bean counters taking their orders–I’m sure nothing would please them more. However, it is rather short-sighted to maintain jobs will be lost on a massive scale. If anything, this should make the hyper-concentrated folks sit up and take notice. It’s no longer simply OK to be an Exchange admin, in fact, it never was and I always looked at these folks with some suspicion.

There are quite a number of operational processes and systems that can simply not be migrated to “the cloud.” As long as those exist, a good sysadmin will continue to have a job. Shields, on the other hand, needs to dust off an elementary logic text.

If say a “cloud service” is cumbersome, slow, or the Systems Planner did not properly analyze “the fit”, then its promise could end up being more costly than the old way.

I’ve witnessed Systems designers who are hell bent on implementing a pet design “to save costs” but in actuality what the end result was that the user organizations were burden with increased costs. Those increased costs were slower performance, clumbsy interfaces, downtime, etc.

Better make sure the Systems Designers think before they act.

I dont think that the applications are ready yet, to totally go cloud based, but i would say in 2-5 years it will be.

At that point in time, acting as a consultant to help decide what cloud based services and how to integrate them into their infrastructure will be what is needed. Oh also physical networking to some degree.