From: Dario Faggioli <dario.faggioli@citrix.com>
To: Wei Liu <wei.liu2@citrix.com>
Cc: "xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>,
甘清甜 <qingtiangan@gmail.com>
Subject: Re: Fwd: Question about the status of vNUMA in Xen
Date: Tue, 15 Sep 2015 17:49:53 +0200 [thread overview]
Message-ID: <1442332193.7789.11.camel@citrix.com> (raw)
In-Reply-To: <20150908095037.GE11097@zion.uk.xensource.com>
[-- Attachment #1.1: Type: text/plain, Size: 1873 bytes --]
On Tue, 2015-09-08 at 10:50 +0100, Wei Liu wrote:
> On Tue, Sep 08, 2015 at 01:21:04AM +0000, 甘清甜 wrote:
> > > So far, I have read the Xen NUMA Roadmap page and watched the
> > > video about vNUMA in Xen on Youtube. I have known that some work
> > > has been done for vNUMA in Xen. Since my benchmark result show a
> > > great improvement in Xen-4.5.1 when compared with Xen-4.0.1. I'm
> > > puzzled if vNUMA has contributed to that improvement.
> >
> > No. That's mostly due to other improvements.
> >
> > Wei.
> > I’m sorry that I didn’t express clearly in the last mail. What I want to
> > know is
> >
> > that how much the NUMA optimizations (including NUMA-aware VM placement,
> >
> > NUMA-aware scheduling and vNUMA) contribute to the improvement
> >
>
Going from 4.0 to 4.5, you get automatic placement and NUMA aware
scheduling, not to mention that many other improvements happened in
between those two release, even if they may not look NUMA-related at a
first glance, may well be leading to better performance.
The NUMA roadmap page is slightly outdated, and I commit to update it
ASAP.
Anyway, if you want to isolate the contribution of each single feature,
you can configure things in such a way that you 'temporarily disable'
them. For instance, you can use vcpu pinning di tweak/disable automatic
placement, and check how the numbers change. For tweaking/disabling NUMA
aware scheduling, you should play with hard and soft affinity (i.e.,
some more refined ways of doing pinning, check what `xl vcpu-pin' does
these days).
Regards,
Dario
--
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
[-- Attachment #2: Type: text/plain, Size: 126 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
prev parent reply other threads:[~2015-09-15 15:49 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-06 7:03 Question about the status of vNUMA in Xen 甘清甜
2015-09-07 10:23 ` Wei Liu
2015-09-08 1:21 ` Fwd: " 甘清甜
2015-09-08 9:50 ` Wei Liu
2015-09-15 15:49 ` Dario Faggioli [this message]
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=1442332193.7789.11.camel@citrix.com \
--to=dario.faggioli@citrix.com \
--cc=qingtiangan@gmail.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xenproject.org \
/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;
as well as URLs for NNTP newsgroup(s).