From: Matthias Welk <welk@fokus.gmd.de>
To: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: linux-kernel@vger.kernel.org
Subject: Re: nfs slowdown since 2.5.4
Date: Tue, 4 Jun 2002 17:56:54 +0200 [thread overview]
Message-ID: <200206041756.55696.welk@fokus.gmd.de> (raw)
In-Reply-To: <200206041253.44446.welk@fokus.gmd.de> <shsg0032pxw.fsf@charged.uio.no>
[-- Attachment #1: Type: text/plain, Size: 1679 bytes --]
On Tuesday 04 June 2002 14:23, Trond Myklebust wrote:
> >>>>> " " == Matthias Welk <welk@fokus.gmd.de> writes:
> > Hi, since 2.5.4 I noticed a big slowdown in nfs. It seems that
> > this is related to the changes in the nfs-lookup code, because
> > now most traffic via nfs is for lookup- and getattr-calls as
> > you can see in the attached tcpdump log. I'v also attached a
> > log of nfsstat, which shows this problem too.
>
> Tough... Those extra checks are needed in order to ensure data cache
> correctness on file open().
>
> If you think you don't need them because the files that you are
> reading are known never to change on the server, you can try mounting
> with the 'nocto' mount option.
>
I tried again with the 'nocto' mount option, although I can't be sure that the
files will not change on the server, but the behaviour didn't change :-(
To get more info about this problem I compared the compile time and the nfs
traffic between 2.4.18-4 and 2.5.20 on a RedHat 7.3 system.
The sources (mosfet-liquid0.9.5.tar.gz - KDE style) were located on the local
disc and the libraries were linked over nfs.
The results attached to this mail show the big difference !
I hope you can do anything with it.
Greetings, Matthias.
--
---------------------------------------------------------------
From: Matthias Welk office: +49-30-3463-7272
FHG Fokus mobile: +49-179- 1144752
Kaiserin-Augusta-Allee 31 fax : +49-30-3463-8672
10589 Berlin email : welk@fokus.fhg.de
----------------------------------------------------------------
[-- Attachment #2: build_2.4.18.txt --]
[-- Type: text/plain, Size: 143 bytes --]
#time to build on a RedHat 7.3 system with 2.4.18-4
real 1m1.301s
user 0m50.920s
sys 0m8.390s
lookup: 2820
getattr: 22
readdir: 96
frag: 66
[-- Attachment #3: build_2.5.20.txt --]
[-- Type: text/plain, Size: 162 bytes --]
#time to build on a RedHat 7.3 system with 2.5.20
real 1m48.509s
user 0m51.320s
sys 0m12.790s
lookup: 287582
getattr: 2546
readdir: 1958
frag: 8465
next prev parent reply other threads:[~2002-06-04 15:57 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-06-04 10:53 Matthias Welk
2002-06-04 12:23 ` Trond Myklebust
2002-06-04 15:56 ` Matthias Welk [this message]
2002-06-04 16:48 ` Trond Myklebust
2002-06-04 17:16 ` Matthias Welk
2002-06-04 17:28 ` Trond Myklebust
2002-06-05 11:50 ` Matthias Welk
2002-06-05 12:16 ` Trond Myklebust
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200206041756.55696.welk@fokus.gmd.de \
--to=welk@fokus.gmd.de \
--cc=linux-kernel@vger.kernel.org \
--cc=trond.myklebust@fys.uio.no \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®