We are in the process of taking some performance metrics on a Hyper-V standalone host to get an idea of how things are performing.
One metric we have been watching is the Disk Queue Length:
This particular Hyper-V host has a four drive RAID 10 using 300GB 15K SAS drives connected to a battery backed Intel RAID Controller.
The snip above was taken while three VMs were online. One of the VMs was SBS 2011 running its OS install routines.
It took a little searching to find some information on just what the right Disk Queue length is.
Close to the bottom of the blog post we find:
Inside the “Physical Disk” counter set I tend to monitor only three things. The first is “Current Disk Queue Length” which gives one an idea of how busy the drives are. The “Current Disk Queue Length” should be around two per drive. If you have a RAID 10 volume with 4+4 (total of 8 drives). Then a queue length of 16 is reasonable. A queue length of 32 might indicate this disk is saturated and is the bottleneck in the system.
Well, since we have a 2+2 RAID 10 volume a queue length of 8 or less would be reasonable. So, our high of 5 turns out to be a good thing.
We will keep monitoring things as we go along to see what happens when this particular solution goes into production.
- Intel Server System SR1695GPRX2AC.
- Intel Xeon Processor X3470, 32GB Kingston ECC, Intel RS2BL040 + BBU, RMM
- 300GB 15K Seagate SAS x 4 in RAID 10
- NIC 0 + 2 = Management Team
- NIC 1 + 3 = vSwitch Team (VMLB)
- NIC 5 = RMM bound.
- Virtual Machines:
- Windows Server 2008 R2 Standard 1 + 1 (host and guest).
- Windows SBS 2011 Guest
- Windows 7 Enterprise Guest
There are 13 users on this network with a custom LoB delivered via RDS RemoteApp to the local users as well as a few satellite locations around the province.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book