RAID 5+0 support

RAID 5+0 support

[govind raj]

Hi all,

We are using EVMS 2.5.4 on Linux 2.6.12.6 kernel version.

We find that kernel modules are available for RAID0, 1, 5, 1+0 as part of 
this kernel. But however, we do not find a similar module available for RAID 
5+0. Can someone advise us of how we would be able to get this support added 
into the kernel? If this is not required as a kernel module, how do we 
create a RAID 5+0 using MD?

Thanks in advance for your help,

Govind

Re: RAID 5+0 support

[Kyle Moffett]

On Jan 19, 2006, at 11:52, govind raj wrote:
> Hi all,
>
> We are using EVMS 2.5.4 on Linux 2.6.12.6 kernel version.
>
> We find that kernel modules are available for RAID0, 1, 5, 1+0 as  
> part of this kernel. But however, we do not find a similar module  
> available for RAID 5+0. Can someone advise us of how we would be  
> able to get this support added into the kernel? If this is not  
> required as a kernel module, how do we create a RAID 5+0 using MD?

Raid math:
RAID(5+0) == RAID(5) + RAID(0)

Commands:

mdadm --create /dev/md0 --level=5 --raid-devices=3 /dev/sda /dev/sdb / 
dev/sdc
mdadm --create /dev/md1 --level=5 --raid-devices=3 /dev/sdd /dev/sde / 
dev/sdf
mdadm --create /dev/md3 --level=0 --raid-devices=2 /dev/md0 /dev/md1

I believe this is all amply documented on a variety of md websites.

Cheers,
Kyle Moffett

-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCM/CS/IT/E/U d- s++: a18 C++++>$ ULBX*++++(+++)>$ P++++(+++)>$ L++++ 
(+++)>$ !E- W+++(++) N+++(++) o? K? w--- O? M++ V? PS+() PE+(-) Y+ PGP 
+ t+(+++) 5 X R? !tv-(--) b++++(++) DI+(++) D+++ G e>++++$ h*(+)>++$ r 
%(--)  !y?-(--)
------END GEEK CODE BLOCK------

Re: RAID 5+0 support

[Phillip Susi]

Why on earth would you want to stripe two raid-5's instead of using one 
raid-5 that is twice as big?  You'd get more usable disk space that way. 


govind raj wrote:
> Hi all,
>
> We are using EVMS 2.5.4 on Linux 2.6.12.6 kernel version.
>
> We find that kernel modules are available for RAID0, 1, 5, 1+0 as part 
> of this kernel. But however, we do not find a similar module available 
> for RAID 5+0. Can someone advise us of how we would be able to get 
> this support added into the kernel? If this is not required as a 
> kernel module, how do we create a RAID 5+0 using MD?
>
> Thanks in advance for your help,
>
> Govind
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe 
> linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>
>

Re: RAID 5+0 support

[Lars Marowsky-Bree]

On 2006-01-19T12:26:10, Phillip Susi <psusi@cfl.rr.com> wrote:

> Why on earth would you want to stripe two raid-5's instead of using one 
> raid-5 that is twice as big?  You'd get more usable disk space that way. 

If redundancy is the goal - to be able to withstand two drive failures
-, RAID6 would be the considerably better choice.


Sincerely,
    Lars Marowsky-Brée

-- 
High Availability & Clustering
SUSE Labs, Research and Development
SUSE LINUX Products GmbH - A Novell Business	 -- Charles Darwin
"Ignorance more frequently begets confidence than does knowledge"

Re: RAID 5+0 support

[Martin Drab]

On Thu, 19 Jan 2006, Phillip Susi wrote:

> Why on earth would you want to stripe two raid-5's instead of using one raid-5
> that is twice as big?  You'd get more usable disk space that way. 

Speed is the issue here, I believe. By stripping two RAID-5 arrays you 
ought to get the reliability of the RAID-5 but with considerably higher 
speed. That's basically why RAID-50 exists, I think.

Martin

Re: RAID 5+0 support

[Phillip Susi]

Martin Drab wrote:
> Speed is the issue here, I believe. By stripping two RAID-5 arrays you 
> ought to get the reliability of the RAID-5 but with considerably higher 
> speed. That's basically why RAID-50 exists, I think.

One big raid-5 would have higher speed because it would have one more 
disk allocated to storing data rather than more parity.  The raid 5+0 
isn't really going to be any more reliable because it can withstand a 
single failure in either half, but not two failures in one half, so in 
the face of a double failure, you have a 50/50 chance of one being in 
each half. 

Re: RAID 5+0 support

[Martin Drab]

On Thu, 19 Jan 2006, Phillip Susi wrote:

> Martin Drab wrote:
> > Speed is the issue here, I believe. By stripping two RAID-5 arrays you ought
> > to get the reliability of the RAID-5 but with considerably higher speed.
> > That's basically why RAID-50 exists, I think.
> 
> One big raid-5 would have higher speed because it would have one more disk
> allocated to storing data rather than more parity.  The raid 5+0 isn't really
> going to be any more reliable because it can withstand a single failure in
> either half, but not two failures in one half, so in the face of a double
> failure, you have a 50/50 chance of one being in each half. 

Well, yes and no. See for instance:

http://en.wikipedia.org/wiki/RAID#RAID_50_.28RAID_5.2B0.29

Martin