Solved

Best Raid config for OS Partition

Posted on 2009-05-07
10
1,477 Views
Last Modified: 2016-12-08
I just got a new HP Proliant server with 8 146 GB 15K drives. I was planning on setting up the first 2 drives as Raid 1, but noticed that they have an option to "Use one drive as spare". Not sure what that means, but is anyone doing Raid 0 and checking the Use drive as spare box?
0
Comment
Question by:judsoncollege
  • 3
  • 3
  • 2
  • +1
10 Comments
 
LVL 1

Accepted Solution

by:
richn earned 300 total points
ID: 24330933
That would be a very bad idea for your OS partition.  RAID 0 means that the data is striped across the drives for performance, but there is no redundancy.  RAID 1 is the way to go.  That way if one drive failes then the other 1 will take over.  Now at that point you are in a delicate situation.  You have lost your redundancy until you replace the defective drive.  That is where the spare comes into play.  If you designate a spare, it will take over for any drive that fails.  Once you replace the failed drive, the controller will rebuild the array and then the spare will go back to being a spare.  One spare drive can cover multiple arrays.  With 8 drives you could go with a RAID 1 for your OS, a RAID 5 with 5 drives for data, and one hot spare.

What role will this server be performing?
0
 

Author Comment

by:judsoncollege
ID: 24331055
It will be a domain controller/app/file server. I would rather have 2 servers, but money is an issue. What is the deal with the Maximum boot partition? Should I disable the Max boot partition? If I do I guess it gives me a 4 GB Max patition and if I Enbale it then it gives 8 GB Max? Thanks. What are your thoughts on leaving leaving one fo the extra 6 drives as a spare? Not sure if I should really waste part of 146 GB.
0
 
LVL 6

Assisted Solution

by:bcoyxp
bcoyxp earned 100 total points
ID: 24331063
hi,

in addition to the above's comment, i'd like to share with you my common practice in an enterprise setup. I am handling a number servers in an enterprise domain and for our standard, RAID 1+0  is applied on first 2 HDDs (you have this option in HP, yet it is actually RAID 1) the OS partition, the rest is set to RAID 5 or RAID 6 (when the number of disks is more than 10)

the advantage of RAID 1+0 is that you could replace the disk (one at a time) "on the fly" on the event of Hard disk failure.

say, i am having same spec of your server; i should be setting it this way:

HDD1 to HDD2 - RAID 1+0 is applied, set for OS partition
HDD3 to HDD8 - RAID 5 is applied, set for Data partition

hope this gives you an idea.

Regards,
0
Optimizing Cloud Backup for Low Bandwidth

With cloud storage prices going down a growing number of SMBs start to use it for backup storage. Unfortunately, business data volume rarely fits the average Internet speed. This article provides an overview of main Internet speed challenges and reveals backup best practices.

 
LVL 6

Expert Comment

by:bcoyxp
ID: 24331128
hi,

have you looked into running Hyper-V instead or any virtualization option?
by this way, you could utilized your server hardware well.
aside from that, application/service conflict shall be prevented.
at least your application server, file server, etc. shall be isolated when a problem occur.

if you are in to windows, you have option to install enterprise edition (with hyper-v). and you can have 4 free licenses for your VM servers.

Regards,  
0
 
LVL 1

Expert Comment

by:richn
ID: 24335550
Bcoyxp, may I ask what RAID 1+0 is?  We are an IBM shop as opposed to HP, and I am unfamiliar with that terminology.
0
 
LVL 26

Assisted Solution

by:lnkevin
lnkevin earned 100 total points
ID: 24335586
Not sure what that means, but is anyone doing Raid 0 and checking the Use drive as spare box?

I will be more specific to your question. NO, not with RAID 0. Basically, spare is using to replace a failed drive and remirror data from the existing ACTIVE drive. If you use RAID 0, when it's down, there is no existing ACTIVE drive to copy the data back so you will loose your data.
For the best OS configuration I will go with the above suggestion:

2X146GB for OS and application (C: partition for OS and D: or E: for all applications and logs)
6X146GB for Fileshare (F: partition).

K
0
 
LVL 26

Expert Comment

by:lnkevin
ID: 24335651
the advantage of RAID 1+0 is that you could replace the disk (one at a time) "on the fly" on the event of Hard disk failure....

Not if your drives are not hot swappable. If your drives are hot swappable you can replace it "on the fly" (while server is running) even if you are running RAID 5. Basically, RAID 1+0 is RAID 1 mirror set applied for more than 2 drives in the even number of drives. That said, you can only set RAID 1+0 (RAID 10) with 4, 6, 8, 10..... drives. RAID 10 is the best write performance wise set up, but it takes the most space for redundancy.

K
0
 
LVL 1

Expert Comment

by:richn
ID: 24336079
If RAID 1+0 is the same as RAID 10, then it has no benefit over RAID 1 from the perspective of being able to replace a failed drive "on the fly"; both RAID levels support that.  From the perspective of performance, your OS drive should never be large enough to require RAID 10.  If it is, then you have too much crap stored on it.
To reinforce what I said in my first post, and what has been repeated by others, RAID 0 gives you no redundancy in the case of a failed drive and is not appropriate for your OS drive, or any other drive where the drive contains the primary copy of important data.  The most common use of RAID 0 would be for something like streaming video, where there is another copy of the data that can then be restored to the RAID 0 array should it fail.  While the RAID 0 array was in a failed state, the data would be unavailable.  If there was no backup copy of the data, it would also be unrecoverable.
0
 
LVL 6

Expert Comment

by:bcoyxp
ID: 24336188
hi,

pardon for the late reply, Inkevin's comment is right with regards to "on the fly" comment.
IBM do have the same practice on RAID systems, you're still be able to apply such system to IBM.
since you have mentioned that an HP machine is what you are having, just make use of the HP smartstart CD provided with. through it you will be guided in deploying your server.

Regards,
0
 

Author Comment

by:judsoncollege
ID: 24379456
Thanks guys.
0

Featured Post

PRTG Network Monitor: Intuitive Network Monitoring

Network Monitoring is essential to ensure that computer systems and network devices are running. Use PRTG to monitor LANs, servers, websites, applications and devices, bandwidth, virtual environments, remote systems, IoT, and many more. PRTG is easy to set up & use.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

INTRODUCTION The purpose of this document is to demonstrate the Installation and configuration, of the HP EVA 4400 SAN Storage. The name , IP and the WWN ID’s used here are not the real ones. ABOUT THE STORAGE For most of you reading this, you …
Moving your enterprise fax infrastructure from in-house fax machines and servers to the cloud makes sense — from both an efficiency and productivity standpoint. But does migrating to a cloud fax solution mean you will no longer be able to send or re…
This video shows how to quickly and easily add an email signature for all users on Exchange 2016. The resulting signature is applied on a server level by Exchange Online. The email signature template has been downloaded from: www.mail-signatures…
Nobody understands Phishing better than an anti-spam company. That’s why we are providing Phishing Awareness Training to our customers. According to a report by Verizon, only 3% of targeted users report malicious emails to management. With compan…

810 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question