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


[Xen-devel] vcpu/cpu

To: xendevel <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] vcpu/cpu
From: jd <jdsw2002@xxxxxxxxx>
Date: Sun, 8 Nov 2009 13:56:36 -0800 (PST)
Delivery-date: Sun, 08 Nov 2009 13:57:02 -0800
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1257717397; bh=PgcujYdW0eH5mgb4grL1cT2b4ocn2iDrUvUs+yc5UFM=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=e3/ou7vhRtqFPTVjb0SnnubgKqik4n3A+CJG0wwNhIBOOG3exxzg8P+3165ij3jfP15W+f7aqqCwfEKHPKx3bKVXOZ4rAEp97RgFvpUwNgaHgOf44vV3wHX2io6SPCCQv6aePihk7xoeVWwVtb0WDYJpacXX/3J1HmYXbzTCrLg=
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=LStJyaFaLuZtOR2LakzOVDKJsoFzxOBlj5GBtTgRggIE/1GcjUCS7f5xRWJZRMfNehvWtdfUx3dlutXn8iyirHnrxaCUKGBsfxM4HACSA0PtYqLAr+fXyn/9gagJ+IETuL6s1C8Ql6o/Jp84QoMYegIi1hl2Sy288KgDlpd4kz8=;
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
    Wanted some clarification on the way xen schedules vcpus on physical cpus.

 - Lets assume there are 8 physical cpus.
   If I set vcpu = 2, for a VM, can it use more that 2 physical cpu time  ? i.e. does VM sees 2 more powerful processors ? Assuming that no-one else need it.

  From the explanation of credit scheduler on the wiki, it seems that the guest kernel would not schedule more that 2 units for work at a time (as it sees 2 cpus). So it looks like the max it can consume is 200%... or max 2 physical cpu time.

  Or the PV driver in the guest does better ?

  Is there any difference between the PV guest and HVM guest in this regard ?

  A related question is : Given the cpu utilization reported by xm top, can I infer cpu utilization within guest (reported by top, or task manager ?).

  Any significant changes in cpu scheduling approach between xen versions.. xen 3.1, 3.2, 3.3, 3.4 ?


Xen-devel mailing list
<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] vcpu/cpu, jd <=