Guidence creating / breaking up disk pools on EMC AX4-5 SAN

Hi All,

I've done some research but I can't seem to find the information I am looking for and decided to utilize this forum to perhaps get some guidance on my issue.  Let me break down some specifics for you all

We have:

EMC Clariion AX4-5 SAN with two storage processors
Two Disk Pools, one holding VMWare data and one holding Sharepoint data

The Disk Pool I am looking to review is the one containing VMWare information.  This Disk Pool currently houses 13 400GB Hard drives (Totaling 4.65TB of data) spanning a RAID 5 Array.  

Does anyone know if there are any guidelines as far as how large a Disk Pool should grow before creating a new disk pool (Rather than expanding to add more drives).  

Data throughput is a concern since at some point the RAID 5 array becomes inefficient not to mention the time it takes to expand an array that size is significant.  Is there any way to measure the amount of throughput to find out at what point the number of drives becomes a bottleneck to the system?  

I have also been tasked with breaking up the current disk pool but I will likely work with EMC get this done.  Any tips in this venture are also appreciated.
RyanMyers83Asked:
Who is Participating?
 
BigSchmuhConnect With a Mentor Commented:
I would consider some numbers:
-Rebuild time : 13x 400GB drives RAID5 requires reading 12x400GB = 4.8TB to rebuild the faulty 400GB
-Risk of a second hdd failure during at rebuild time : most large Enterprise class HDD has a UBE of 1 per 10e15 which means they have about 0.9% a of an unreadable sector per 1TB read...a bit more than 4% probability in your 13x drives case !

SAN vendors are supposed to extensively test their HDD models and even has some agreement with drive manufacturers to modify the drive firmware.
==> Those SAN vendors should provide you their own guidelines. If they can not, you have mine above ;-)
0
 
RyanMyers83Author Commented:
Looks like we are going to break this up a few different ways but it will be a project to figure out how we are going to do that.  I appreciate the guidance and yes the risk of having two drives go is higher with that many drives and needs to be addressed.  It may be a few months before we deploy a solution on this (many different parties are affected and we need a project plan in place) but if I remember I will come back and post our solution and breakdown.
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.