On Fri, May 7, 2010 at 2:06 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;">
Hello,<div><br></div><div>I have a sparse linear system, block dense, one block for each process, and solve with cg and block jacobi (ICC(0)) as preconditioners, with different reordering. During my tests I found quasi linear speed on kspsetup() and KSPSetUpOnBlocks() using 1wd,nd,rcm reorderings, but using qmd I get superlinear speedup. My CPU is a Xeon 5410.</div>
<div><br></div><div>Is that possible or something is going wrong?, and if I want to explain that, what functions on this stage should I monitor??. I really </div><div><span style="font-family:arial, sans-serif;line-height:25px">appreciate</span> any advice. </div>
</blockquote><div><br></div><div>It is possible (but unlikely) that the ordering on a larger number of processes creates a better</div><div>preconditioner, particularly because ICC(0) is so unpredictable. You can try and separate</div>
<div>arithmetic efficiency from algorithmic efficiency by looking at the number of iterates between these</div><div>runs.</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;">
<div>Thanks in advance.</div><div><div><br>-- <br>Pedro Torres<br>GESAR/UERJ<br>Rua Fonseca Teles 121, São Cristóvão<br>Rio de Janeiro - Brasil<br>
</div></div>
</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>