This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


RE: [Xen-ia64-devel] Time for hybrid virtualization?

To: Kayvan Sylvan <kayvan@xxxxxxxxx>
Subject: RE: [Xen-ia64-devel] Time for hybrid virtualization?
From: tgingold@xxxxxxx
Date: Fri, 25 Jan 2008 11:36:30 +0100
Cc: Alex Williamson <alex.williamson@xxxxxx>, xen-ia64-devel <xen-ia64-devel@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Fri, 25 Jan 2008 02:36:45 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <DF0031A8C0149E41802BAFD18D6C4B701294E6370B@xxxxxxxxxxxxxxxxxxxxxx>
List-help: <mailto:xen-ia64-devel-request@lists.xensource.com?subject=help>
List-id: Discussion of the ia64 port of Xen <xen-ia64-devel.lists.xensource.com>
List-post: <mailto:xen-ia64-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ia64-devel>, <mailto:xen-ia64-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ia64-devel>, <mailto:xen-ia64-devel-request@lists.xensource.com?subject=unsubscribe>
References: <1200528127.6773.88.camel@lappy><1200592688.7657.72.camel@bling> <20080118062441.GA2564@saphi> <1200699825.7133.32.camel@lappy> <51CFAB8CB6883745AE7B93B3E084EBE201670748@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <1200994146.4795b762d3648@xxxxxxxxxxx> <1201025874.6751.44.camel@lappy> <1201179494.47988b665cbbd@xxxxxxxxxxx> <DF0031A8C0149E41802BAFD18D6C4B701294E636F1@xxxxxxxxxxxxxxxxxxxxxx> <1201207442.6826.49.camel@lappy> <DF0031A8C0149E41802BAFD18D6C4B701294E6370B@xxxxxxxxxxxxxxxxxxxxxx>
Sender: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Internet Messaging Program (IMP) 3.2.8
Quoting Kayvan Sylvan <kayvan@xxxxxxxxx>:

> For the Montecito chip, running the RE-AIM7 compute workload, we have the
> following preliminary summary:
> Native vs. Xen performace
> ----------------------------------
> At 2-CPU, the overhead was 13%.
> At 4-CPU, the overhead was 17%
> At 8-cpu, overhead was 28%
> At 16-cpu, the overhead was 173%
> Not quite sure why the performance dropped off so radically in the 16-CPU
> case.

Can you tell us a little bit more about your machine?

Is it 16 logical cpus (hyperthread) or 16 cores ?

Are all the cpus on the same bus or is it a NUMA machine ?

That could explain the drop off.

Xen-ia64-devel mailing list