<div dir="ltr"><div class="gmail_extra"><br><div class="gmail_quote">On Sat, Mar 2, 2013 at 1:17 AM, Jie Chen <span dir="ltr"><<a href="mailto:jiechen@mcs.anl.gov" target="_blank">jiechen@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div id=":4bh">For the former case, you are not trying to squeeze one subdomain to one compute node, are you? Why is it memory limited?<br>
</div></blockquote><div><br></div><div style>That's what I had in mind. Yes, you can use subdomains much larger than one process, but it's easy to use smaller subdomains, and then the Galerkin coarse operator becomes somewhat sparse.</div>
<div style><br></div><div style>Is there a demonstration of deflation providing significant benefit for problems that are _not_ amenable to multigrid?</div></div></div></div>