This is the situation. 60 odd users with Outlook 2003 running in online mode (a few use caching mode). File & Print, DC, Exchange all on one physical box (as separate VM's) connected over 100Mbs About 15 miles away) over a LAN with servers and users all in the same subnet. We would get the occasional performance problem (Outlook is trying to retrieve data from the Exchange Server) especially when opening many calendars at once but usually it's fine. Then the service provider moved File & Print and Exchange into a different subnet (doing a VM Ware move so the machines still have the same configuration). Now we are experiencing many more instances of the retrieve data messages and often Outlook will hang waiting for Exchange to respond. Checking the VM running Exchange doesn't show high CPU usage so it's somewhat puzzling. The SP has told us, we could do two things. We should try to move users to caching mode to ease load on the network and the server. We will do that when it's appropriate. But they also told us we should implement a local DC and I am getting conflicing advice as to what sort of machine that should be. One piece of advice is, any old server lying around should do the trick, another says we should get a server grade machine with good CPU and memory. I have a Blackberry server running Windows Server 2003 SBS supporting only 6 users. It's a HP desktop (core2 duo with 2G RAM) and I thought I could press that into service if a local DC is in fact the solution. But what puzzles me is, the performance issue only became significant once the Exchange server was moved into a different subnet (albeit on the same network fabric it was on before) and since we can communicate with it most of the time, it doesn't seem to be a network configuration error. Just wondering if forum members have any ideas on what might be the best way forward here. Thanks Larry

What is the disk performance on your Exchange Server? That is usually the case. I would say for sure put your users in Cached mode. Mark Morowczynski|MCT| MCSE 2003:Messaging, Security|MCITP:EMA 2K7,EDA Win 7,ES,SA,EA|MCTS:Windows Mobile Admin|Security+|http://almostdailytech.com

Any update on your issue? Have you checked the disk performance like Mark wrote and tried Cached mode? Also check the underlying virtulization "OS" so it can provide the necessary I/O's Jonas Andersson MCTS: Microsoft Exchange Server 2007/2010 | MCITP: EMA 2007/2010 | MCSE/MCSA Blog: http://www.testlabs.se/blog

Hi The outsourcer who host our systems looked at disk space on the servers and found them the drives tbe quite severely fragmented. So they defragged the drives. However it didn't really make any difference since users are still reporting the same problems as before. They have also recommended we go to caching mode and have created a GP and a script for us to use. That is we defined which users we want to migrate, move them into the GPO and then have them restart or reboot. The area of concern I have here is, we have a number of users who mailboxes > 2G with a few in 4G and up to 10G range. I guess we could exclude them from the caching mode or move them after hours to reduce network choking. But what really concerns me is the problems started to occur once the Exchange Server was moved from one physical machine (that was running Exchange, a File & Print service and SQL) on the same subnet as our PC's to a new VM on a different physical server in a different subnet. They told me they just VM Moved the server which of course moved it warts and all (hence the fragmentation on a new server instance) What in the move could have triggered this problem? In my view a local DC or caching mode might help but it seems to be fixing the symptoms, not the problem.