[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-users] workings of live migration

  • To: xen-users@xxxxxxxxxxxxxxxxxxx
  • From: Thomas Hofkens <thomas.hofkens@xxxxxxxxx>
  • Date: Thu, 4 Jun 2009 15:28:47 +0200
  • Delivery-date: Fri, 05 Jun 2009 04:11:50 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type :content-transfer-encoding; b=IV8AI0r4a3t0DrRO79fPP15/Oi390EbzkYTqkvYnLcx1NzkwZRet4ndtPQTMoAXout HjDO/56O4fkusZhUbkQJCOpEqj90kJHVvSPQITnM0Py5MFhgd/8AaqijfRq9md+LGcqM 4S90lKE8zBS6dAjl2cdAGaA0uh3SP/74cBwok=
  • List-id: Xen user discussion <xen-users.lists.xensource.com>


I am researching XenServer from Citrix and (with 2 colleagues)
comparing it with VMware ESX and Microsoft HyperV.

In our tests, it seems that Xen's live migration is using less
resources than VMware's ESX and I would like to know why that is. I
found an article from last year that references a paper from 2005,
explaining what actually happens with the pages during live migration.

This is an extract of that article about the memory transfer:

>Push phase- The source VM continues running while certain pages are pushed 
>across the network to the new
>destination. To ensure consistency, pages modified during this process must be 
>Stop-and-copy phase The source VM is stopped, pages are copied across to the 
>destination VM, then the new VM is >started.
>Pull phase The new VM executes and, if it accesses a page that has not yet 
>been copied, this page is faulted in
>("pulled") across the network from the source VM.

I was wondering if the memory transfer still happens in the same
fashion as it did 4 years ago.

Thank you

Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.