HP Raid 5 rebuild calculation on Proliant cluster

I am trying to calculate how long the rebulid time will be on my clusters array.
I am expanding the array by upgrading the disks from 176 gb to 300gb drives
the drives are 10k rpm with 64 kb stipe size.
The array has 7 drives and is 683 gb in size and 95% full
The Raid controller has 128mb Ram cache

Current rebuild priority low.

Please let me know your predictions (I will do this out of hours with low disk IO)
FrencheeeAsked:
Who is Participating?
 
andyalderConnect With a Mentor Commented:
36 hours is quite believable, assuming it's done during a quiet period then rebuild priority doesn't matter much, if server I/O is high increeasing rebuild priority can slow down the server. There's a white paper on rebuild rate Vs performance at http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00419664/c00419664.pdf (MSA500 controller is same technology as 6402).

It would be so much faster if you had RAID 10, take about 20 minutes per disk.
0
 
fm250Commented:
I rebuit mine last time which took more than the estamated time 24 hours, but was 1TB. So I would advise you to do the following:
1- have your systems ghosted -images-
2- get all your data backed up
3 - make sure you do it during the  weekend or holdays - it may take extra time- so notify your staff.
4- updates all the firmware and drivers for any related hardware
hope this helps!
0
 
FrencheeeAuthor Commented:
How long did it take per drive? And what size were your old drives?
0
Improve Your Query Performance Tuning

In this FREE six-day email course, you'll learn from Janis Griffin, Database Performance Evangelist. She'll teach 12 steps that you can use to optimize your queries as much as possible and see measurable results in your work. Get started today!

 
fm250Commented:
250 gb each drive. the whole process took more than 24 hrs to rebuild.
0
 
andyalderCommented:
The standard rebuild time is 4GB/hour for HP smart array controllers (and the MSA500 is effectively a smart array controller). On a cluster though it may not rebuild at all if the quorum is involved since it gets accessed continually so there is no quiet time for rebuild to run. Unfortunately the only fix for this is to stop the cluster service and hence bring the servers down effectively.
0
 
FrencheeeAuthor Commented:
The Quorum drive is on another array mirror on 72gb drives so no problem there. Is that 146/4 = 36.5 Hours!! What about if I increase the rebuild priority.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.