The problem started on Friday when the internet speeds on our site had slowed to a crawl at best. Late that afternoon I had tracked the problem to our GMS server which at that time was using 4Mb/s of up traffic sustained. I restarted the server and the problem appeared to be resolved.

However, upon arriving Monday morning I found the server to be in the same condition. I once again restarted the server only this time I stopped all GMS services. I waited several hours and the problem did not return (ruling out anything in the OS itself). I then re-started all GMS services, and the problem re-turned within roughly 15-20 min.

I have watched the dashboard graphs closely and do not see any one user or multiple users who have any in/out traffic blasting (at least according to their graphs).
Also none of the system graphs appear to have anything sustained going on.

One thing that is interesting is that the problem will only start once GMS services have been enabled, however stopping the services after does not resolve the condition. Up traffic remains through the roof at that point.

Our site is only synching 21 devices.

Any help at all in narrowing this down would be greatly appreciated.