Terminal Server Slowdown

1 Host Server running HyperV, Server 2008R2 Core
4 live VMs all stored on 1 separate physical disk (separate from the boot drive)
The DC is virtualized on this server as well
Roaming profiles are enabled

A Terminal Server VM has about 35 users and doesn't have any performance issues throughout the day (light use: firefox, office etc.) but in the morning when many users are logging in, the system has serious performance issues.

Any thoughts on what I can do to improve performance?
Who is Participating?
The is likely two fold.
1) your profiles are stored on the same drive as the drive to which the roaming profiles is being written.
2) All your VMs are on the same disk so that further concentrates the resource dependency on disk access. When 10 users login at the same time, the roaming profile needs to retrieved from the fileserver VM on the same disk as the TerminalServer VM where the data is then being stored.  Other disk,  does perfmon reflect an increase in Page file hits/misses? Which could also be triggering VM disk access/translating into physical disk access.
Have user logins staggered.
It sounds that within a window of time all users login which might require a large amount of profile data import.

Do you alo use folder redirection?  It might help reduce the amount of data that needs to be transferred at logon.
lineonecorpAuthor Commented:
Would leaving users logged in help with this scenario? I'm not sure we'd be able to stagger the users.

We do use folder redirection and it is hosted by the DC on the same HyperV server.
Free Tool: Port Scanner

Check which ports are open to the outside world. Helps make sure that your firewall rules are working as intended.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

Are your synchronization settings set for logon or logoff.
When a user logs into a session, what is happening? How long does the logon process take?
One option is to use a tool such as cacti to collect resource usage using snmp data polling.
CPU, memory, network, etc.
Using that info you might be in a better position to see where the issue might be, you can use perfmon.
All your cms resideon the same separate disk, his might be where the issue is.
lineonecorpAuthor Commented:
Log on during peak access (15 - 20 users in 20 minutes) takes about 3 minutes.

Log on normally is less than 20 seconds

I've been using Perfmon and have noticed the disk queue of between 8 and 12 during peak times. Would moving the DC to a separate disk help resolve this?
lineonecorpAuthor Commented:
Still responding?
lineonecorpAuthor Commented:
Thanks for keeping up with this and sorry for being so slow to get back. I will try rejigging the disks and get back to you with the results.
lineonecorpAuthor Commented:
Worked like a charm.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.