From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Andres Lagar-Cavilla" Subject: Re: Cloning a VM and copy-on-write deduplicating memory using CoW page sharing in Xen 4+ Date: Thu, 2 Feb 2012 11:23:47 -0800 Message-ID: References: Reply-To: andres@lagarcavilla.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Patrick Wilbur Cc: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org > Hey all, > Hey Andres, > > I'm looking to clone a VM into several extremely-similar VMs, and I'm > hoping to also make use of your lovely new CoW page sharing capabilities > in > Xen 4. > > From my understanding of a previous thread where Andres described the > process of sharing/coalescing memory between VMs, it sounds like I will > need to "manually" coalesce each page using a homebrew tool of mine. The > issue I have with doing this is it seems like I'd need to pause, save mem, > load mem in a new VM, coalesce, and resume two VMs, which seems painful > and > wasteful of a process for cloning! Patrick, that is indeed painful and wasteful. That is why we added xc_memshr_add_to_physmap. Now you can do Potemkin-/SnowFlock-like cloning in a few lines of code. You still want the source VM to be paused, obviously. Yuengling is my favourite US beer ;) Andres > > Is there an easier way to do this, or should we add a new feature for CoW > cloning of VMs in Xen via a userspace tool? > > Thanks, > Pat Wilbur & team > > > -- > Patrick F. Wilbur > Researcher, Consultant, Educator, > Computer Science Graduate at Clarkson University > > DONE RIGHT THE FIRST TIME: Consulting and hiring information: > http://pdub.net/consulting/ & http://pdub.net/hiring/ > > patrick.wilbur@gmail.com > wilburpf@clarkson.edu > > Check out our book: http://runningxen.com > My website: http://pdub.net >