You're so bored at work that you gotta stir a pot that didn't need stirring? :)
That being said, I can see what you're aiming for. Instead of having 3 distribution points that a client will try #1, fail, try #2, fail, try #3. You can have 3 distribution points all working in tandem to send the data as requested, and if one of those servers isn't reachable for whatever reason, the other 2 will keep on truckin' without interruption or yucky failures in the logs.
Does that sound about right?
