mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Serge E. Hallyn" <serue@us.ibm.com>
To: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: "Serge E. Hallyn" <serue@us.ibm.com>,
	linux-kernel@vger.kernel.org, dev@sw.ru, herbert@13thfloor.at,
	devel@openvz.org, sam@vilain.net, xemul@sw.ru,
	Dave Hansen <haveblue@us.ibm.com>, Andrew Morton <akpm@osdl.org>,
	Cedric Le Goater <clg@fr.ibm.com>
Subject: Re: [PATCH 0/9] namespaces: Introduction
Date: Sun, 21 May 2006 11:27:59 -0500	[thread overview]
Message-ID: <20060521162759.GA19707@sergelap.austin.ibm.com> (raw)
In-Reply-To: <m1sln61jqs.fsf@ebiederm.dsl.xmission.com>

Quoting Eric W. Biederman (ebiederm@xmission.com):
> "Serge E. Hallyn" <serue@us.ibm.com> writes:
> 
> > This patchset introduces a per-process utsname namespace.  These can
> > be used by openvz, vserver, and application migration to virtualize and
> > isolate utsname info (i.e. hostname).  More resources will follow, until
> > hopefully most or all vserver and openvz functionality can be implemented
> > by controlling resource namespaces from userspace.
> >
> > Previous utsname submissions placed a pointer to the utsname namespace
> > straight in the task_struct.  This patchset (and the last one) moves
> > it and the filesystem namespace pointer into struct nsproxy, which is
> > shared by processes sharing all namespaces.  The intent is to keep
> > the taskstruct smaller as the number of namespaces grows.
> 
> 
> Previously you mentioned:
> > BTW - a first set of comparison results showed nsproxy to have better
> > dbench and tbench throughput, and worse kernbench performance.  Which
> > may make sense given that nsproxy results in lower memory usage but
> > likely increased cache misses due to extra pointer dereference.
> 
> Is this still true?  Or did our final reference counting tweak fix
> the kernbench numbers?
> 
> I just want to be certain that we don't add an optimization,
> that reduces performance.

Here are the numbers with the basic patchsets.  But I guess I should
do another round with adding 7 more void*'s to represent additional
namespaces.

(intervals are for 95% CI, tests were each run 15 times)

           |  with nsproxy  |   without nsproxy |
kernbench  | 68.90 +/- 0.21 |   69.06 +/- 0.22  |
dbench     | 386.0 +/- 26.6 |   388.4 +/- 21.0  |
tbench     | 391.6 +/- 8.00 |   389.4 +/- 10.95 |

reaim with nsproxy
1 115600.000000 5512.441557
3 246985.712000 9375.780582
5 272309.092000 8029.833742
7 290020.000000 7288.367116
9 298591.580000 5557.531915
11 nan nan
13 nan nan
15 nan nan

reaim without nsproxy
1 110160.000000 5728.697311
3 246985.712000 9375.780582
5 262204.197333 11138.510652
7 288660.000000 6880.898412
9 300631.580000 4351.926692
11 nan nan
13 nan nan
15 nan nan

  parent reply	other threads:[~2006-05-21 16:28 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-18 15:47 Serge E. Hallyn
2006-05-18 15:48 ` [PATCH 1/9] namespaces: add nsproxy Serge E. Hallyn
2006-05-21 23:30   ` Sam Vilain
2006-05-21 23:38     ` Eric W. Biederman
2006-05-22 12:39       ` Serge E. Hallyn
2006-05-18 15:49 ` [PATCH 2/9] namespaces: incorporate fs namespace into nsproxy Serge E. Hallyn
2006-05-18 15:49 ` [PATCH 3/9] namespaces: utsname: introduce temporary helpers Serge E. Hallyn
2006-05-18 15:49 ` [PATCH 4/9] namespaces: utsname: switch to using uts namespaces Serge E. Hallyn
2006-05-19  0:02   ` Randy.Dunlap
2006-05-19  2:21     ` Serge E. Hallyn
2006-05-19  2:45       ` Randy.Dunlap
2006-05-19  3:12       ` Sam Vilain
2006-05-19  9:05     ` Eric W. Biederman
2006-05-19 17:39       ` Randy.Dunlap
2006-05-19 11:58     ` Eric W. Biederman
2006-05-22 19:43     ` Cedric Le Goater
2006-05-22 20:19       ` Randy.Dunlap
2006-05-22  0:19   ` Sam Vilain
2006-05-18 15:49 ` [PATCH 5/9] namespaces: utsname: use init_utsname when appropriate Serge E. Hallyn
2006-05-18 15:50 ` [PATCH 6/9] namespaces: utsname: implement utsname namespaces Serge E. Hallyn
2006-05-18 15:50 ` [PATCH 7/9] namespaces: utsname: sysctl hack Serge E. Hallyn
2006-05-18 15:50 ` [PATCH 8/9] namespaces: utsname: remove system_utsname Serge E. Hallyn
2006-05-18 23:03   ` Paul Mackerras
2006-05-18 23:04     ` Paul Mackerras
2006-05-18 15:51 ` [PATCH 9/9] namespaces: utsname: implement CLONE_NEWUTS flag Serge E. Hallyn
2006-05-18 17:34 ` [PATCH 0/9] namespaces: Introduction Andrew Morton
2006-05-18 19:23   ` John Kelly
2006-05-18 23:28   ` Sam Vilain
2006-05-18 23:43     ` Sam Vilain
2006-05-19  4:24     ` Paul Jackson
2006-05-19  9:23       ` Eric W. Biederman
2006-05-19 11:41   ` Eric W. Biederman
2006-05-19 17:52     ` Jeff Dike
2006-05-20  0:16     ` Sam Vilain
2006-05-19 12:42   ` Herbert Poetzl
2006-05-19 15:13     ` Andrew Morton
2006-05-19 16:27       ` Eric W. Biederman
2006-05-19 16:40         ` Andrew Morton
2006-05-19 17:15           ` Stephen Hemminger
2006-05-19 20:17           ` Dave Hansen
2006-05-19 20:52             ` Alexey Kuznetsov
2006-05-19 18:28         ` Hua Zhong
2006-05-19 19:38           ` Serge E. Hallyn
2006-05-19 19:45           ` John Kelly
2006-05-19 20:23             ` John Kelly
2006-05-19 20:04       ` Dave Hansen
2006-05-20  3:18         ` Eric W. Biederman
2006-05-21  0:48         ` Eric W. Biederman
2006-05-21 22:57       ` Pavel Machek
2006-05-21 23:18         ` Eric W. Biederman
2006-05-21 23:32           ` Herbert Poetzl
2006-05-22 16:54             ` Eric W. Biederman
2006-05-19 13:47   ` Andrey Savochkin
2006-05-19 15:25     ` Andrew Morton
2006-05-20 21:24       ` Herbert Poetzl
2006-05-22 17:23       ` Eric W. Biederman
2006-05-20  0:16     ` Sam Vilain
2006-05-19  8:50 ` Eric W. Biederman
2006-05-19 13:30   ` Serge E. Hallyn
2006-05-21 16:27   ` Serge E. Hallyn [this message]
2006-05-21 18:08     ` Eric W. Biederman
2006-05-22 12:10       ` Serge E. Hallyn
2006-05-22 16:44         ` Eric W. Biederman
2006-05-19 17:17 Al Boldi

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=20060521162759.GA19707@sergelap.austin.ibm.com \
    --to=serue@us.ibm.com \
    --cc=akpm@osdl.org \
    --cc=clg@fr.ibm.com \
    --cc=dev@sw.ru \
    --cc=devel@openvz.org \
    --cc=ebiederm@xmission.com \
    --cc=haveblue@us.ibm.com \
    --cc=herbert@13thfloor.at \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sam@vilain.net \
    --cc=xemul@sw.ru \
    /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®