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-research] Re: Motivation?

To: "Y. DU" <duyuyang@xxxxxxxxx>
Subject: [Xen-research] Re: Motivation?
From: Andres Lagar-Cavilla <andreslc@xxxxxxxxxxxxxx>
Date: Thu, 25 Sep 2008 14:12:23 -0400
Cc: xen-research <xen-research@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Thu, 25 Sep 2008 11:09:23 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <200809251008212479230@xxxxxxxxx>
List-help: <mailto:xen-research-request@lists.xensource.com?subject=help>
List-id: Research Issues on Xen <xen-research.lists.xensource.com>
List-post: <mailto:xen-research@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-research>, <mailto:xen-research-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-research>, <mailto:xen-research-request@lists.xensource.com?subject=unsubscribe>
References: <200809251008212479230@xxxxxxxxx>
Sender: xen-research-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Thunderbird (X11/20070530)
http://compbio.cs.toronto.edu/snowflock actually redirects to
http://sysweb.cs.toronto.edu/projects/1. Maybe something in your setup
cannot handle redirects?

Your questions are very open-ended. We could rant for ages. I suggest
you read our technical report instead, where we outline some good
reasons for doing what we're doing.

One important thing is that our starting point is cloud or shared
clusters that rely on virtualization. There are many well-known good
reasons to do that: isolation, security, ease of administration, etc.
>From that starting point, what is a good method to come up with VMs very
fast, very cheaply, when you need them to, say, handle bursts? That's
one place where SnowFlock is very helpful.


Y. DU wrote:
> http://compbio.cs.toronto.edu/snowflock.
> Page Not Found
> Precisely, what is the benefit from spreading computation on the fly
> to distribution/parallelism?
> And, in the unit of VM, pros outweigh cons?
> Shawn.
> 2008-09-25
> ------------------------------------------------------------------------
> Y. DU

Xen-research mailing list

<Prev in Thread] Current Thread [Next in Thread>