so basically what your saying is something along the lines of a rendering cluster would be a good candidate for this?<br><br><div class="gmail_quote">On Wed, Jan 27, 2010 at 6:31 PM, Jon Forrest <span dir="ltr"><<a href="mailto:jlforrest@berkeley.edu">jlforrest@berkeley.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">At a recent Rocks clustering user's group<br>
meeting the recent addition of Rocks support of<br>
Xen-based virtual clusters came up. Some<br>
of the same questions recently raised on this<br>
list were discussed there.<br>
<br>
One justification for virtual clusters that I<br>
hadn't thought of was discussed. This only applies<br>
in places with large clusters run by a central<br>
computing group but used by various internal<br>
customers. Using virtual clusters makes it<br>
very easy to supply clusters to customers<br>
who need a cluster for a limited period of<br>
time. The amount of effort necessary to<br>
provision a new cluster is minimal.<br>
Nodes can easily and quickly be added,<br>
if necessary. This is as opposed to buying<br>
a new cluster for a research group, using it<br>
for a couple of months, and then turning it<br>
off.<br>
<br>
So, in this case, virtualized clusters have<br>
the advantage of being easier to manage. The<br>
performance overhead caused by the virtualization<br>
is a factor, but it's decreasing as time goes<br>
on due to better hardware support of virtualization<br>
and cleverer software.<br>
<br>
Cordially,<br>
-- <br>
Jon Forrest<br>
Research Computing Support<br>
College of Chemistry<br>
173 Tan Hall<br>
University of California Berkeley<br>
Berkeley, CA<br>
94720-1460<br>
510-643-1032<br>
<a href="mailto:jlforrest@berkeley.edu" target="_blank">jlforrest@berkeley.edu</a><div><div></div><div class="h5"><br>
_______________________________________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
To change your subscription (digest mode or unsubscribe) visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" target="_blank">http://www.beowulf.org/mailman/listinfo/beowulf</a><br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>Jonathan Aquilina<br>