Dell/Perc5 raid/MPT SAS Integrated Raid Write Performance

[Filip Palian]

Tom Judge pisze:
> Filip Palian wrote:
>> Hi folks,
>>
>> I've also encountered problems with write performnce on PE860, below are
>> some details.
>>
>>
> 
> I believe that a fix for this went into RELENG_6 this week.
> Scott Long MFC'd the following fix at 2007-06-05 21:32:57.
> 
> Tom
> 

Great thing that sombody cares.


> Commit Message:
> 
> scottl      2007-06-03 23:13:05 UTC
> 
>   FreeBSD src repository
> 
>   Modified files:
>     sys/dev/mpt          mpt.c mpt.h mpt_cam.c
>   Log:
>   mpt.c:
>   mpt.h:
>           Add support for reading extended configuration pages.
>   mpt_cam.c:
>           Do a top level topology scan on the SAS controller.  If any
> SATA device are discovered in this scan, send a passthrough FIS to set
> the write cache.  This is controllable through the following tunable at
> boot:
>     hw.mpt.enable_sata_wc:
>           -1 = Do not configure, use the controller default
>            0 = Disable the write cache
>            1 = Enable the write cache
> 
>           The default is -1.  This tunable is just a hack and may be
>           deprecated in the future.
>

Unfortunatly I couldn't use this fix because disk performance is crucial
for me and i could not depend on the "hack" which may be "deprecated in
the future".

> Turning on the write cache alleviates the write performance problems
> with SATA that many people have observed.  It is not recommend for those
> who value data reliability!  I cannot stress this strongly enough.
> However,  it is useful in certain circumstances, and it brings the
> performence in line with what a generic SATA controller running under
> the FreeBSD ATA driver provides (and the ATA driver has had the WC
> enabled by default for years).
> 

Just for curiosity I've tried to run FreeBSD 6.2 in Xen and KVM with
Gentoo as a host system. Under Xen nothing could be done because of "BTX
halted" error with the same values as described in the bug
"http://lists.freebsd.org/pipermail/freebsd-bugs/2006-November/020906.html"
(in my example it was an IDE cdrom). But happily KVM did the work. I've
intsalled FreeBSD and run blogbench and guess what... The results were:

Final score for writes: 84
Final score for reads : 13636

Summing up, I'll wait for 7.0 release and then I'll put my hands on it.

Regards,
Filip Palian