<p>PCASM with multiple subdomains per process.</p>
<p>PCGAMG coarse levels.</p>
<div class="gmail_quote">On Jan 25, 2012 12:37 PM, "Matthew Knepley" <<a href="mailto:knepley@gmail.com">knepley@gmail.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Wed, Jan 25, 2012 at 12:29 PM, Dominik Szczerba <span dir="ltr"><<a href="mailto:dominik@itis.ethz.ch" target="_blank">dominik@itis.ethz.ch</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
>> I recently realized that, independently of my explicitly partitioning<br>
>> the input mesh, Petsc also employs partitioning somewhere internally.<br>
><br>
><br>
> We don't unless you tell us to.<br>
<br>
Can you please expand? I am partitioning my input unstructured mesh to<br>
distribute dofs throughout the processes, and then I am setting up my<br>
MPI matricess "as usual", not intending any calls to parmetis or other<br>
partitioning functions... So where can I tell or not tell Petsc to use<br>
partitioning here or not?<br></blockquote><div><br></div><div>Unless you create a MatPartitioning object, we do not call ParMetis.</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Many thanks<br>
<span><font color="#888888">Dominik<br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>
</blockquote></div>