March 29, 2004, 12:00 AM — Last week we talked about inter-site replication - or replication
between sites. This week, we look at why we have sites in the first
place and what their role is in Active Directory (AD).
A site is defined as one or more domain controllers (DCs) connected with
high network bandwidth. By defining sites within AD, you effectively
tell AD what your underlying physical network looks like. This allows
DCs to utilize the underlying network in the most efficient manner
possible - and allows AD to conserver bandwidth that is required for
other applications within your organization.
In addition to giving DCs the information they require to replicate data
efficiently within your network, sites also allow clients efficient
access to domain resources. By default, when a client logs onto the
domain, the PC will try to find a domain controller within the clients'
home site. If one cannot be found, then an attempt will be made to find
a DC in another site. This usually speeds up the logon process for the
user, since they are utilizing DCs within their own site. Additionally,
it minimizes logon traffic that would potentially be routed across slow
If you recall in NT 4.0 domains, it was possible that one day you would
logon to the domain and be authenticated by a DC in your home location,
and then the next day you may be authenticated by a DC clear around the
world. This was not always the case, but there was no good way to
control this situation. With Windows 2000 Active Directory (or Windows
2003 AD for that matter), this is possible by using sites effectively.
Join me next week when we talk about the knowledge consistency checker
(KCC) and intra-site replication.