Date: Thu, 29 Nov 2007 18:07:04 +0100 From: Dominic Fandrey <LoN_Kamikaze@gmx.de> To: Norberto Meijome <freebsd@meijome.net> Cc: freebsd-stable@freebsd.org Subject: Re: msdosfs performance unbearable Message-ID: <474EF1B8.1030606@gmx.de> In-Reply-To: <20071128225517.30a8b26a@meijome.net> References: <474BD99C.7070002@gmx.de> <20071128225517.30a8b26a@meijome.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Norberto Meijome wrote:
> On Tue, 27 Nov 2007 09:47:24 +0100
> Dominic Fandrey <LoN_Kamikaze@gmx.de> wrote:
>
>> ufs:
>> $ time -h tar -xf php_manual_en.tar.gz
>> 3.31s real 0.43s user 0.51s sys
>
> I've seem something similar , in the past, on 6.2, when writing to my mobile phone's mini-SD card.
>
> what does gstat show? (in particular, is any device being used 100% ?, can u relate the slowness when it hits 100% ? do other disks other than your FAT disk become saturated too? )
This is the gstat output on UFS for reference:
dT: 1.006s w: 1.000s
L(q) ops/s r/s kBps ms/r w/s kBps ms/w %busy Name
151 2106 47 1416 12.3 2059 11425 21.4 60.8| ad0
0 0 0 0 0.0 0 0 0.0 0.0| acd0
0 0 0 0 0.0 0 0 0.0 0.0| ad0s1
151 2092 33 525 12.1 2059 11425 21.8 53.5| ad0s2
0 0 0 0 0.0 0 0 0.0 0.0| ad0s3
0 14 14 891 12.8 0 0 0.0 17.8| ad0s4
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2a
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2b
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2c
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2d
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2e
151 2092 33 525 12.1 2059 11425 22.5 54.3| ad0s2f
0 0 0 0 0.0 0 0 0.0 0.0| cd0
And this is the same operation on fat32:
dT: 1.017s w: 1.000s
L(q) ops/s r/s kBps ms/r w/s kBps ms/w %busy Name
1 11 0 0 0.0 11 43 0.1 0.2| ad0
0 0 0 0 0.0 0 0 0.0 0.0| acd0
0 0 0 0 0.0 0 0 0.0 0.0| ad0s1
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2
0 0 0 0 0.0 0 0 0.0 0.0| ad0s3
1 11 0 0 0.0 11 43 0.2 0.2| ad0s4
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2a
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2b
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2c
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2d
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2e
0 0 0 0 0.0 0 0 0.0 0.0| ad0s2f
0 0 0 0 0.0 0 0 0.0 0.0| cd0
As you can see, there is a /slight/ difference in throughput. Both slices are
on the same HD, so it's not a controller thing.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?474EF1B8.1030606>