Firstly thanks for your patience all, we know its frustrating, and trust me we would rather all was running smoothly as well!
The current issue seems to be caused when we ramped up the number of new machines supporting the dimensions (who would have though more processing power was a bad thing eh?) and that simply didn't show up in our scale testing when we tested the hardware.
Now, on the plus side the coders and the server team now believe they have located the problem in the network and are currently working on patching the relevant network drivers on the new machines so that we can be up as soon as possible.
The bad news is that will take a couple more hours, but I can assure you that we are all working as hard as we can to resolve this issue.