• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Nvidia Raid 0+1 Information

M

Mindflux

Member
I've been googling around, but can't seem to find a solid description.

I realise Nvidia's 0+1 requires 4 drives. 2 for striping and two for mirroring. However here is the question.

Are the mirrored drives just mirrors of disk0 and 1's stripe? Or is it a mirror of the complete data (times two)?

Does that make sense?

Disk 0 - Stripe
Disk 1 - Stripe
Disk 2 - Mirror of Disk 0
Disk 3 - Mirror of Disk 1

or

Disk 0 - Stripe
Disk 1 - Stripe
Disk 2 - Mirror of both striped drives (into functioning single disk)
Disk 3 - Mirror of both striped drives (or mirror of Disk2)

 
Ok that's what I was figuring.

I've got high disc queue length on my MSSQL server even though nobody was on it this weekend. So I'm either going to set up Raid 0+1 to lower it, or keep my current mirror (Raid 1), and toss in two Raptors as Raid 0 and move the MDF(database) to the Raptors and then back up the MDF nightly to the mirror.

Ultimatey I need to get the disk queue avg lower than what it's at (4.337 steady, with no users.. which is a bit of an oddball).


 
0.o I think you could get more results if you fixed your config first. I don't see how you can have a queue length with no requests.
 
Yeah I'm not entirely sure either. I can see the perfmon show IO spikes now and then and with process explorer it looks like csrss is causing the disk I/O.

 
You must log in or register to reply here.

TRENDING THREADS

Back
Top