On Fri, Mar 26, 2010 at 6:44 PM, Pedro Torres <span dir="ltr"><<a href="mailto:torres.pedrozpk@gmail.com">torres.pedrozpk@gmail.com</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;">
Hi, <br><br>I'm evaluating how to handle differents mapping and ordering, and I would like to have some concept of how expensive will be to put it in a loop. Is it possible to order these functions according the computational cost??.<br>
<pre><font size="4">AOApplicationToPetsc()</font><br><font size="4">AO</font><font><font size="4">Petsc</font></font><font><font size="4">To</font></font><font size="4">Application()</font><br><br><font size="4">ISGlobalToLocalMappingApply</font><font size="4">()</font><br>
<font size="4">ISLocalToGlobalMappingApply</font><font size="4">()</font></pre></blockquote><div><br></div><div>These are all just direct array lookups to an index translation table. The AO have a somewhat high memory</div>
<div>cost in parallel, but the performance should be near identical for all of these.</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;">
<pre><br><font size="4">Thanks you very<br><br><br>Pedro</font><br><br></pre><br>
</blockquote></div><br><br clear="all"><br>-- <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>