A common volume design for servers running the 'Failover Clustering' component is to utilize mount points instead of exhausting the available 26 drive letters. The clustering software allows you to define volume dependencies to ensure root volumes are online prior to mount points, I am curious as to the boot order for non-clustered servers with regards to drives and mount points as there doesn't seem to be a way to define volume dependencies for those volumes assigned as mount points.
Drives:
C: OS
D: Data
E: pagefile
F: mount root #1
G: mount root #2
Mounts:
F:\mounts\volume5
F:\mounts\volume6
F:\mounts\volume7
F:\mounts\volume8
G:\mounts\volume9
G:\mounts\volume10
G:\mounts\volume11
G:\mounts\volume12
The overall goal is to provide a drive letter for each SQL instance installed on a consolidated server allowing for secondary volumes (data1,data2,logs,tempdb,backup,etc). Since this is a consolidated SQL server hosting multiple development SQL instances, since it is DEV performance isn't a major concern, but I'd like to configure individual volumes to cut down on disk queuing and any related disk latency which would occur if we combined SQL data,log,tempdb datafiles.
Any and all thoughts appreciated !!
Thanks,
-chadwic