Due to our mailbox database sizes becoming very high, I was going to come on here and ask peoples opinions on how the split there organizations databases down (ie. do people split there databases by department or do they just randomly drop there users into say mdb1, mdb2, mdb3 etc etc?) We currently split our datbases by department but due to the size of these databases, it is becomming hard to manage and obvious that we will need to change our current database structure - any thoughts on this would be greatly apprecicated. However, when checking one our Mailbox Servers today, I noticed that one particular mailbox server is only showing as 7gb of Free Disk space out of 500gb. This is on the data drive, we have split our log files onto a different drive. We are planning to grow the data drive by 100gb today, but my question is that even though it shows the disk usage on this MB Server on as nearly 493gb - when you look into the mailbox databases underneath them, they only total about 363gb there are no hidden folders or files. Can someone please advise on what would cause this - is this database fragmentation, white space?

With regards to the Free Disk space we have found that this was due to the System Volume Information folder. So not really Exchange related http://www.symantec.com/business/support/index?page=content&id=TECH83223 However, I would still like peoples thoughts on how they split there mailbox databases down. Do they go for a departmental database model, or find that this often quickly makes the database sizes unmanagable.

I have done every method there is. What it comes down to is something that is predictable and allows the data to be evenly spread. The most common is surname. Take the list of users and mailbox sizes, count each letter and spread them out. If it gets uneven, more the users around. Simon.Simon Butler, Exchange MVP Blog | Exchange Resources | In the UK? Hire Me.

Thank you Simon