Miserable performance of cyrus-imapd 2.3.9 -- seems to be lockingissues

[Simon Matter]

> On Tue, 4 Mar 2008, Ian G Batten wrote:
>
>>  software RAID5 is a performance
>> disaster area at the best of times unless it can take advantage of
>> intimate knowledge of the intent log in the filesystem (RAID-Z does
>> this),
>
> actually, unless you have top-notch hardware raid controllers, software
> raid 5
> may be better then hardware raid 5. many controllers only do a decent job
> doing
> raid 0 or raid 1. this is something to measure with your particular
> hardware.
> I've seen many cases where the cards do a horrible job with raid 5
> compared to
> software.

I can only second that. I'm still wondering what "top-notch hardware raid
controllers" are. From my experience the only decent "controllers" you can
get are those in the heavy priced SAN equipments with gigs of cache on the
SAN controllers and tens or hundreds of spindles behind it.

>
>> For a terabyte, 3x500GB SATA drives in a RAID5 group will be blown
>> out of the water by 4x500GB SATA drives in a RAID 0+1 configuration
>> in terms of performance and (especially) latency, especially if it
>> can do the Solaris trick of not faulting an entire RAID 0 sub-group
>> if one spindle fails.  Rebuild still isn't pretty, mind you.

You can partly implement the trick of not faulting an entire drive by
splitting the drive into smaller parts. With a 500G drive you can split it
into 10 segments of 50G and create 10 RAID(1 or 5) devices on it. Just put
thos 10 RAID segments into the same volume group. In case of some single
sector defects it will only degrade 10% of your entire RAID and you could
easily move those PV's to a spare RAID1 disk before touching any disk.

Simon