I have a 100 user network that are very heavy users. The main site is 50 users and the 2 sites are 25 users each. The connection between the main site and the remote sites is 100mb fiber. Each site is on their own subnet. I currently have a DC at each remote site providing GC, DNS and DHCP. I have an RDS application that each user at the remote sites access that is hosted at the main site.
I am currently being challenged on why I have DCs at the remote sites. I have been doing this since 2000. I have always put DCs at remote sites for those roles (and DFS-r sometimes). I know in my MSCE 2003 studies it was definitely best practice to do this. Have things changed? Does anyone have some best practice information that supports this in either direction?