MD1000-PERC5/E Performance issues

[Mike McGrath]

On Fri, 2 Jan 2009, Matthias Saou wrote:

> Mike McGrath wrote :
>
> > On Sun, 28 Dec 2008, Mike McGrath wrote:
> >
> > > I've read up on:
> > > http://thias.marmotte.net/archives/2008/01/05/Dell-PERC5E-and-MD1000-performance-tweaks.html
> > >
> > > Which seems to have helped but I really don't think I'm seeing the
> > > performance data I should be.  Take the attached sar data.  I started by
> > > doing a recursive cp of some files from the filesystem to some other
> > > drives, then started the sar dump.
> > >
> > > At marker 12:12:55, I started to do a cat /dev/sde > /dev/null.  (/dev/sde
> > > is the array).
> > >
> > > Why was there such a significant jump there?  Also, why does sar think the
> > > array was 100% utilized basically the whole time even though there's such
> > > a difference in sec reads/s?
> > >
> > > Other issues.  /dev/sde is the underlying physical device (raid5).
> > > However, it is divided up using lvm.  The copy mentioned above was from
> > > a logical volume.  While I expected some overhead and performance hit, I
> > > didn't expect it to be so huge and sporadic.  I've seen reads as low as
> > > 3MB/s during a copy.
> > >
> > > Thoughts?
> > >
> >
> > Ok, upon further testing I don't think the majority of my issue is related
> > to the MD1000 or the raid controller.  I'm seeing decent read speeds from
> > large files.  So I'll get back to debugging unless someone has a tip.  I
> > suspect another list would be better.
>
> Do post back your findings here, though, as you're far from being the
> first to face performance issues with the MD1000... ;-)
>

Will do, there's some further discussion here:

https://www.redhat.com/archives/fedora-infrastructure-list/2008-December/msg00142.html

and

https://www.redhat.com/archives/fedora-infrastructure-list/2009-January/msg00000.html

	-Mike