Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 2 Oct 2009 10:29:50 -0700 (PDT)
From:      Richard Mahlerwein <mahlerrd@yahoo.com>
To:        freebsd-performance@freebsd.org
Subject:   Re: A specific example of a disk i/o problem
Message-ID:  <160343.45338.qm@web51006.mail.re2.yahoo.com>

next in thread | raw e-mail | index | archive | help
>From: Dieter <freebsd@sopwith.solgatos.com>=0A>Subject: Re: A specific exa=
mple of a disk i/o problem=0A>To: freebsd-performance@freebsd.org=0A>Date: =
Friday, October 2, 2009, 1:07 AM=0A>=0A>Updated demo, just to make sure:=0A=
>=0A># big_file is larger than main memory, on same disk as man (/usr)=0A>t=
ime man de=A0=A0=A0 =A0=A0=A0 =0A>cat big_file > /dev/null=A0=A0=A0=0A>cat =
big_file > /dev/null &=A0=0A>time man de=0A=0ATrying it on something that s=
urely will have i/o issues:=0APIII 500 w/ 60 GB PATA drive and 256 MB RAM.=
=A0 Blank=0Aminimal install + man pages right now.=A0 I'm trying with=0Aa "=
file /etc/" command instead of trying to reliably time=0Aman... It seems th=
e file command always hits the disk.=A0=0AThis should also make any sort of=
 "flushing" unecessary,=0Ashouldn't it?  Anyway, I tried it with the man co=
mmand but couldn't discern any difference at all with it, so I'm repeating =
with "file" which at least gives SOME variance.=0A =0Aliebnitz# uname -a=0A=
FreeBSD xyz 7.2-RELEASE FreeBSD 7.2-RELEASE #0: Fri May=A0 1 08:49:13 UTC 2=
009=A0 =A0=A0=A0root@walker.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC=A0=
=0Ai386=0A=0Aliebnitz# time dd if=3D/dev/zero of=3D/usr/bigfile count=3D409=
6 bs=3D1M=0A4096+0 records in=0A4096+0 records out=0A4294967296 bytes trans=
ferred in 363.700967 secs (11809062 bytes/sec)=0A0.055u 83.971s 6:03.78 23.=
0%=A0 =A0 25+1071k 128+34391io 0pf+0w=0A=0Aliebnitz# time file /etc/ [run 5=
 times]=0A4.052u 0.047s 0:04.14 98.7%=A0 =A0=A015+1085k 0+0io 0pf+0w=0A4.06=
8u 0.031s 0:04.13 99.0%=A0=A0=A0=A016+1095k 0+0io 0pf+0w=0A4.052u 0.046s 0:=
04.13 99.0%=A0=A0=A0=A016+1101k 0+0io 0pf+0w=0A4.044u 0.054s 0:04.13 99.0%=
=A0=A0=A0=A016+1092k 0+0io 0pf+0w=0A4.052u 0.046s 0:04.13 99.0%=A0=A0=A0=A0=
15+1091k 0+0io 0pf+0w=0A=0AOk, so that looks pretty repeatable.=A0 Now...=
=0A=0Aliebnitz# cat /usr/bigfile > /dev/null &=0A[1] 37923=0Aliebnitz# time=
 file /etc/* [again run 5 times]=0A4.146u 0.038s 0:06.80 61.3%=A0=A0=A0=A01=
5+1103k 87+0io 8pf+0w=0A4.128u 0.054s 0:06.73 61.9%=A0=A0=A0=A015+1095k 93+=
0io 6pf+0w=0A4.127u 0.056s 0:06.65 62.7%    14+1058k 88+0io 4pf+0w=0A4.111u=
 0.072s 0:06.84 61.1%    15+1049k 91+0io 6pf+0w=0A4.115u 0.066s 0:06.86 60.=
7%    15+1022k 91+0io 6pf+0w=0A=0ABTW, after the cat is done...=0Aliebnitz#=
 time file /etc/*=0A4.051u 0.047s 0:04.13 99.0%     15+1078k 0+0io 0pf+0w=
=0A=0AI see no unexpected issues, only a minor slowdown, and that could be =
processer contention, IMO.  =0A=0A(Sorry for duplicate, I was rewriting wit=
h the change to "file" instead of "man" and the two year old decided to pre=
ss some keys and sent the message.)=0A=0A=0A      



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?160343.45338.qm>