Btrfs RAID-10 performance

Miloslav Hůla miloslav.hula at gmail.com
Tue Sep 15 12:44:26 EEST 2020


Dne 15.09.2020 v 10:22 Linda A. Walsh napsal(a):
> On 2020/09/10 07:40, Miloslav Hůla wrote:
>> I cannot verify it, but I think that even JBOD is propagated as a 
>> virtual device. If you create JBOD from 3 different disks, low level 
>> parameters may differ.
> ----
>     JBOD allows each disk to be seen by the OS, as is.  You wouldn't
> create JBOD disk from 3 different disks -- JBOD would give you 3 separate
> JBOD disks for the 3 separate disks.

Yes. If I create 3 JBOD configurations from 3 100GB disks, I get 3 100GB 
devices in OS. If I create 1 JBOD configuration from 3 100GB disks, I 
get 1 300GB device in OS.

>     So for your 16  disks, you are using 1 long RAID0?  You realize
> 1 disk goes out, the entire array needs to be reconstructed.  Also
> all of your spindles can be tied up by long read/writes -- optimal speed
> would come from a read 16 stripes wide spread over the 16 disks.

No. I have 16 RAID-0 configurations from 16 disks. As I wrote, there was 
no other option of how to propagate 16 disks as 16 devices into OS few 
years before.

>     What would be better, IMO, is going with a RAID-10 like your subject
> says, using 8-pairs of mirrors and strip those.  Set your stripe unit
> for 64K to allow the disks to operate independently.  You don't want
> a long 16-disk stripe, as that's far from optimal for your mailbox load.
> What you want is the ability to have multiple I/O ops going at the same
> time -- independently.  I think as it stands now, you are far more likely
> to get contention as different mailboxes are accessed with contention
> happening within the span, vs. letting each 2 disk mirror potentially doing
> a different task -- which would likely have the effect of raising your
> I/O ops/s.

The reason to not create RAID-10 in controller was, that btrfs scrubbing 
detects slowly degrading disk much sooner than controller (verified many 
times). And if I create RAID-10 in controller, btrfs scrub detects soon 
too, but I'm not able to recognize on which disk.

>     Running raid10 on top of raid0 seems really wasteful

I'm not doing that.



More information about the dovecot mailing list