Thats awesome. 80% true efficiency for 1K nodes on under<br>
1 million unknowns. Thats only 1,000 unknowns per proc. If<br>
we were cranking the problems size, I am sure you'd be up<br>
in the 90s. Hopefully this summer we can deliver you much<br>
improved serial performance as well. BTW, I want to talk<br>
sometime about the subduction benchmark. Van Keken talked<br>
about it at Purdue, so I got a diametrically opposed viewpoint<br>
from a sympathetic person. It was eye opening.<br>
<br>
Matt<br><br><div><span class="gmail_quote">On 4/2/06, <b class="gmail_sendername">Richard Katz</b> <<a href="mailto:katz@ldeo.columbia.edu">katz@ldeo.columbia.edu</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi All,<br><br>Thought you might be interested to see this.<br><br>The simulation is finite-volume, steady-state, non-Newtonian CFD in 3D.<br>And, of course, it is done using PETSc (version 2.3.1).<br><br>How does this compare to other PETSc-based simulations?
<br><br>Cheers<br>Rich<br><br>Barry -- I had a problem with the the jobs that we discussed but I'm<br>hoping to run them properly tonight.<br><br><br><br></blockquote></div><br><br clear="all"><br>-- <br>"Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness