Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 4 Aug 2012 14:27:58 +0100
From:      Steve O'Hara-Smith <steve@sohara.org>
To:        freebsd-questions@freebsd.org
Subject:   ZFS bonnie puzzlement
Message-ID:  <20120804142758.0126fded.steve@sohara.org>

next in thread | raw e-mail | index | archive | help
	Hi,

	I've been looking at the performance of my new NAS box - built
using a Jetway JNF99FL-525 with 4GB of 1066 DDR3 and a pair of 2TB Samsung
F4 drives in a mirror. It all works but the performance reports are puzzling

	Running bonnie -s 8192 - reports character and block write times of
around 66MB/s which is consistent with what zpool iostat -v and systat -v
are showing me. Read performance OTOH is strange, zpool and systat both
reporting consistently an aggregated read speed of around 120MB/s during
the block read tests (which seems a bit slow for the drives - and indeed
systat reports the drives at less than 50% utilisation) but bonnie is only
reporting 35MB/s, I see similar discrepancies with simple dd block reads
to /dev/null, in which case my stopwatch agrees with dd.

	I made some more bonnie tests on the boot SSD - there everything is
as expected systat and bonnie agree and the disc gets pushed to 90-100%
utilisation with write speeds over 100MB/s and read speeds over 250MB/s.

	Also a dd of the raw drive (read only of course) runs the drives up
to around 130MB/s easily - dd and systat agree about the speed.

	First up can anyone explain the discrepancy between bonnie and
zpool/systat for the XFS pool ? Secondly can anyone suggest a reason that I
can't seem to get the drives above 50% reported utilisation on reads ?

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:>WIN                                      | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120804142758.0126fded.steve>