March 28, 2008 at 8:47 am
I have read some of the posts about the proper RAID configuration for SQL Server 2005, and they don't seem to address my specific question. The general consensus, I think, seems to be to put the OS on a mirrored C: volume, the database on a RAID 10 configured D: volume, and the log files on on a RAID 10 volume as well. So, each RAID 10 configurations requires at least 4 volumes, and the mirrored set requires 2 volumes. My new server only has 6 drives in it. When I configure the RAID 10 configuration, I can allocate the 4 physical drives into 2 separate volumes that SQL will recognize as 2 separate drives. Then I can put the database on one, and the log files on the other. From the SQL perspective, they would be 2 separate drives, but physically they would be on the same drives. I will get redundancy on both, but I suspect that performance will suffer. If anyone can shed any light on this, it would be much appreciated.
March 28, 2008 at 8:54 am
I'd say you're right. IMO - it doesn't "buy" you anything to have multiple volumes in the same RAID set. Performance-wise, you haven't changed the number of spindles, so you're just setting the up to "compete" with each other no matter what. If you were to leave things to auto-grow/shrink, i suppose that it might help with reducing fragmentation, but since doing a lot of auto-growing is a bad thing anyway - I would set things up to avoid that.
Personally - I'd set them up as one single large volume, all the while going back and begging for an external enclosure, a controller and 5 more disks (a hot spare). You can always move data files or logs once the new space arrives.
----------------------------------------------------------------------------------
Your lack of planning does not constitute an emergency on my part...unless you're my manager...or a director and above...or a really loud-spoken end-user..All right - what was my emergency again?
Viewing 2 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply