<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="Generator" content="Microsoft Exchange Server">
<!-- converted from text --><style><!-- .EmailQuote { margin-left: 1pt; padding-left: 4pt; border-left: #800000 2px solid; } --></style>
</head>
<body>
<div>I think there are options to velocity extrapolation that we can try before tackling load-balancing that. Level-set methods have so many different little tricks it's not even funny.
<div><br>
</div>
<div> - Åsmund</div>
<br>
Jed Brown <jed@jedbrown.org> skrev:<br>
</div>
<font size="2"><span style="font-size:10pt;">
<div class="PlainText">Åsmund Ervik <asmund.ervik@ntnu.no> writes:<br>
<br>
> Thanks for the answer, sorry for the long delay here, I got knocked<br>
> out by the flu. Oh well.<br>
<br>
And I'm typing with one hand...<br>
<br>
> From your questions I now realize that perhaps the worst-scaling part<br>
> of our code will be velocity extrapolation for level-set<br>
> advection. This takes some time (not as much as the Poisson equation,<br>
> maybe 5% of that) and only happens in a band around the interface. I<br>
> guess we should look into alternatives to velocity extrapolation.<br>
<br>
And it sounds like this part won't be load-balanced without significant<br>
concessions and effort.<br>
</div>
</span></font>
</body>
</html>