We are running Windows Server 2012 R2 with the Hyper-V role installed. I have discovered that the host's volume C: get's fragmented very quickly (about 4% in only 1h).
Our SCOM configuration triggers an alarm after hitting 25% fragmentation - and this happens, even when using the default scheduled defragmentation interval of once per week. The virtual disk files for the VM running on top are located on a different drive. The disk queue length for disk 0 and disk 1 is more or less 0.1. The most active processes with disk activity are:
- System
- Explorer.exe
- svchost.exe (netsvcs)
- HealthService.exe (SCOM Agent)
- mc2.exe (HP PowerProtector)
- sqlservr.exe (part of our backup solution)
But non of them is higher then 130 KB/sec...
This is what the storage layout looks like:
Disk 0 (RAID1 on 2x disks, strip-size 256 KiB / 256 KiB):
- Volume System Reserved
- Volume OS (C:)
- Volume SPOOL (E:)
Disk 1 (RAID10 on 8x disks, strip-size 256 KiB / 1 MiB):
- Volume DATASTORE (D:)
The RAID controller is a Smart Array P420i. Any advice and input on this? Thanks in advance.
