From: Håvard Bjerke <Havard.Bjerke@idi.ntnu.no>
To: "Mark A. Williamson" <mark.williamson@cl.cam.ac.uk>
Cc: xen-devel@lists.sourceforge.net,
Ian Pratt <Ian.Pratt@cl.cam.ac.uk>,
H?ard Bjerke <Havard.Bjerke@idi.ntnu.no>,
Rune Johan Andresen <Rune.Johan.Andresen@idi.ntnu.no>
Subject: Re: Xen & I/O in clusters - problems! New Information
Date: Thu, 28 Oct 2004 13:51:38 +0200 [thread overview]
Message-ID: <20041028115138.GN23334@idi.ntnu.no> (raw)
In-Reply-To: <200410271940.27680.mark.williamson@cl.cam.ac.uk>
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=unknown-8bit, Size: 2521 bytes --]
We're getting these results in domain 0, not in a VM. As I understand, BVT or apropos scheduling do not apply in domain 0?
Håvard
On Wed, Oct 27, 2004 at 07:40:27PM +0100, Mark A. Williamson wrote:
> Just a thought: you are using the BVT scheduler, right? We haven't tested
> performance with the other schedulers recently but we know something goes
> wrong for IO intensive domains on Atropos.
>
> Mark
>
> On Wednesday 27 Oct 2004 19:26, Ian Pratt wrote:
> > > On Fri, Oct 22, 2004 at 06:47:35PM +0100, Ian Pratt wrote:
> > > > > We tried to compile xen0 with CONFIG_E1000_NAPI = y and got the samre
> > > > > results between
> > > > > two xen dom0 nodes. I am not sure if these interrupts tells anything:
> > > >
> > > > It's the different rate at which the eth0 interrupt counts go up
> > > > during your bandwidth tests that is interesting. e.g. poll it
> > > > once a second during the test.
> > >
> > > We tried sending 1 MB and measured:
> >
> > 1MB isn't really very much with a 128KB socket buffer. Do you get
> > the same results with larger transfers?
> >
> > > non-SMP native Linux:
> > > ~ 130k interrupts
> > > 114 kB/s
> > > native Linux with compiled-in SMP support, single CPU:
> > > ~ 140k interrupts
> > > 114 kB/s
> > > Xen0:
> > > ~ 180k interrupts
> > > 80 kB/s
> >
> > It's pretty odd that Xen's taking more interrupts. Are you using
> > the same native kernel version as you are for Xen?
> >
> > Also, what happens if you boot xen with 'nosmp' on the Xen
> > command line.
> >
> > We've got Xen tcp performance results from a bunch of machines,
> > and dom0 to dom0 performance has always been almost identical to
> > native for 1500 byte MTU packets.
> >
> >
> > Ian
> >
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by:
> > Sybase ASE Linux Express Edition - download now for FREE
> > LinuxWorld Reader's Choice Award Winner for best database on Linux.
> > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/xen-devel
-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_idU88&alloc_id\x12065&op=click
next prev parent reply other threads:[~2004-10-28 11:51 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-10-13 5:12 plan 9 status Ronald G. Minnich
2004-10-13 12:14 ` Mark A. Williamson
2004-10-13 22:18 ` Ronald G. Minnich
2004-10-14 0:52 ` Kip Macy
2004-10-14 8:10 ` Keir Fraser
2004-10-14 10:35 ` Wesley Parish
2004-10-14 11:04 ` Keir Fraser
2004-10-14 15:32 ` Ronald G. Minnich
2004-10-14 15:57 ` Keir Fraser
2004-10-21 16:06 ` Keir Fraser
2004-10-14 15:53 ` Mark A. Williamson
2004-10-14 17:45 ` Xen & I/O in clusters - problems! Rune Johan Andresen
2004-10-14 18:17 ` Ian Pratt
2004-10-14 23:24 ` Mark A. Williamson
2004-10-15 13:48 ` Rune Johan Andresen
2004-10-21 19:12 ` Xen & I/O in clusters - problems! New Information Rune Johan Andresen
2004-10-21 20:24 ` Ian Pratt
2004-10-22 16:36 ` Rune Johan Andresen
2004-10-22 17:47 ` Ian Pratt
2004-10-27 17:11 ` Håvard Bjerke
2004-10-27 18:26 ` Ian Pratt
2004-10-27 18:40 ` Mark A. Williamson
2004-10-28 11:51 ` Håvard Bjerke [this message]
2004-10-28 13:09 ` Mark A. Williamson
2004-10-28 11:32 ` Håvard Bjerke
2004-10-28 12:44 ` Ian Pratt
2004-10-28 16:45 ` Håvard Bjerke
2004-10-29 17:05 ` Xen & I/O in clusters - Single Vs. Dual CPU issue Rune Johan Andresen
2004-10-29 17:24 ` Ian Pratt
2004-10-29 18:39 ` Håvard Bjerke
2004-10-29 19:46 ` Ian Pratt
2004-11-02 16:13 ` Håvard Bjerke
2004-11-02 16:35 ` Mark A. Williamson
2004-11-02 16:51 ` Mark A. Williamson
2004-11-04 17:42 ` Rune Johan Andresen
2004-11-04 18:41 ` Bin Ren
2004-11-05 14:53 ` Rune Johan Andresen
2004-11-05 16:47 ` Keir Fraser
2004-11-05 17:33 ` Rune Johan Andresen
2004-11-06 8:04 ` Keir Fraser
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=20041028115138.GN23334@idi.ntnu.no \
--to=havard.bjerke@idi.ntnu.no \
--cc=Ian.Pratt@cl.cam.ac.uk \
--cc=Rune.Johan.Andresen@idi.ntnu.no \
--cc=mark.williamson@cl.cam.ac.uk \
--cc=xen-devel@lists.sourceforge.net \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.