This is a collection of benchmark runs with the FFSB test suite. The
results are comparable to other raid results on the boxacle server.
The latest results can be found at:
http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5.html
I added XFS without barriers and EXT3 with barriers into the mix since
my last data set.
In general things look about the same. There are a few cases where ext4
with barriers (default) has gotten better but ext4 without barriers has
gotten worse.
Thanks,
Keith Mannthey
LTC Local FileSystem
On Fri, Aug 06, 2010 at 12:07:13PM -0700, Keith Mannthey wrote:
> This is a collection of benchmark runs with the FFSB test suite. The
> results are comparable to other raid results on the boxacle server.
>
> The latest results can be found at:
> http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5.html
>
> I added XFS without barriers and EXT3 with barriers into the mix since
> my last data set.
Thanks, it's good to know there haven't been any major regression on
the ext4 side.
If you could have a chance to run a set of benchmark runs with the
jbd2 scalability patches, I'd be hugely interested to see how much
difference they make.
Thanks again!
- Ted
On Fri, 2010-08-06 at 15:14 -0400, Ted Ts'o wrote:
> On Fri, Aug 06, 2010 at 12:07:13PM -0700, Keith Mannthey wrote:
> > This is a collection of benchmark runs with the FFSB test suite. The
> > results are comparable to other raid results on the boxacle server.
> >
> > The latest results can be found at:
> > http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5.html
> >
> > I added XFS without barriers and EXT3 with barriers into the mix since
> > my last data set.
>
> Thanks, it's good to know there haven't been any major regression on
> the ext4 side.
>
> If you could have a chance to run a set of benchmark runs with the
> jbd2 scalability patches, I'd be hugely interested to see how much
> difference they make.
Yes I am running with those today. As the system is only a 16-way or so
it will be interesting to see the impact.
I also have a few barrier reduction patchset test runs as well. I will
work to put them all together and publish some ext4 only relevant run in
the next few days.
Thanks,
Keith
> Thanks again!
>
> - Ted
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html