The fact that Exchange 2010 includes database availability groups has some implications that you may not have thought of. Most of them, I hadn’t thought of either, which is why it’s great that there are smart people like Ross Smith IV (the original; accept no substitutes). Ross sent some suggested practices for Exchange 2010 database naming to TAP admins, and I wanted to share them.
Here’s the deal: in Exchange 2007 and earlier, mailbox and public folder databases are children of server objects. That means that you can uniquely identify a database by a combination of its name (which may not be unique throughout the forest) and its server name (which is guaranteed by AD to be unique). In Exchange 2010, the database is no longer “owned” by a particular server. Instead, it’s a member of a DAG, and it may actually become active on any server in the DAG at any time. That means that your database names shouldn’t include the name of the server. DAGs can span AD sites, too, so guess what: don’t use the AD site name (or the name of the physical datacenter) either. Otherwise the name of the database may not correspond in any way to where the database is actually active.
Finally, consider carefully whether you want to include the name of the organization or company. This has nothing to do with DAGs per se, but rather with the overhead of updating database names after a merger, acquisition, or rebranding. Unless, of course, you work for Contoso, in which case you should be OK.
Paul's Down-Home Page 