<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, Feb 28, 2018 at 9:01 PM, TAY wee-beng <span dir="ltr"><<a href="mailto:zonexo@gmail.com" target="_blank">zonexo@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<br>
<div class="m_8518435943966620320moz-cite-prefix">On 1/3/2018 12:10 AM, Matthew Knepley
wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">On Wed, Feb 28, 2018 at 10:45 AM, TAY
wee-beng <span dir="ltr"><<a href="mailto:zonexo@gmail.com" target="_blank">zonexo@gmail.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
I have a CFD code which uses PETSc and HYPRE. I found that
for a certain case with grid size of 192,570,048, I
encounter scaling problem when my cores > 600. At 600
cores, the code took 10min for 100 time steps. At 960,
1440 and 2880 cores, it still takes around 10min. At 360
cores, it took 15min.<br>
<br>
So how can I find the bottleneck? Any recommended steps?<br>
</blockquote>
<div><br>
</div>
<div>For any performance question, we need to see the output
of -log_view for all test cases.</div>
</div>
</div>
</div>
</blockquote>
Hi,<br>
<br>
To be more specific, I use PETSc KSPBCGS and HYPRE geometric
multigrid (entirely based on HYPRE, no PETSc) for the momentum and
Poisson eqns in my code.<br>
<br>
So can log_view be used in this case to give a meaningful? Since
part of the code uses HYPRE?<br></div></blockquote><div><br></div><div>Make an event to time the HYPRE solve. It only takes a few lines of code.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div text="#000000" bgcolor="#FFFFFF">
I also program another subroutine in the past which uses PETSc to
solve the Poisson eqn. It uses either HYPRE's boomeramg, KSPBCGS or
KSPGMRES.<br>
<br>
If I use boomeramg, can log_view be used in this case?<br></div></blockquote><div><br></div><div>Yes, its automatic.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div text="#000000" bgcolor="#FFFFFF">
Or do I have to use KSPBCGS or KSPGMRES, which is directly from
PETSc? However, I ran KSPGMRES yesterday with the Poisson eqn and my
ans didn't converge.<br></div></blockquote><div><br></div><div>Plain GMRES is not good for Poisson. You would be better off with GMRES/GAMG.</div><div><br></div><div> Thanks,</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 text="#000000" bgcolor="#FFFFFF">
Thanks.<br>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I must also mention that I partition my grid only in the x
and y direction. There is no partitioning in the z
direction due to limited code development. I wonder if
there is a strong effect in this case.</blockquote>
<div><br>
</div>
<div>Maybe. Usually what happens is you fill up memory with
a z-column and cannot scale further.</div>
<div><br>
</div>
<div> Thanks,</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"><span class="m_8518435943966620320HOEnZb"><font color="#888888"><br>
-- <br>
Thank you very much<br>
<br>
Yours sincerely,<br>
<br>
==============================<wbr>==================<br>
TAY Wee-Beng 郑伟明 (Zheng Weiming)<br>
Personal research webpage: <a href="http://tayweebeng.wixsite.com/website" rel="noreferrer" target="_blank">http://tayweebeng.wixsite.com/<wbr>website</a><br>
Youtube research showcase: <a href="https://www.youtube.com/channel/UC72ZHtvQNMpNs2uRTSToiLA" rel="noreferrer" target="_blank">https://www.youtube.com/channe<wbr>l/UC72ZHtvQNMpNs2uRTSToiLA</a><br>
linkedin: <a href="http://www.linkedin.com/in/tay-weebeng" rel="noreferrer" target="_blank">www.linkedin.com/in/tay-weeben<wbr>g</a><br>
==============================<wbr>==================<br>
<br>
</font></span></blockquote>
</div>
<br>
<br clear="all"><span class="HOEnZb"><font color="#888888">
<div><br>
</div>
-- <br>
<div class="m_8518435943966620320gmail_signature" data-smartmail="gmail_signature">
<div dir="ltr">
<div>
<div dir="ltr">
<div>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</div>
<div><br>
</div>
<div><a href="http://www.caam.rice.edu/%7Emk51/" target="_blank">https://www.cse.buffalo.edu/~<wbr>knepley/</a><br>
</div>
</div>
</div>
</div>
</div>
</font></span></div>
</div>
</blockquote>
<br>
</div>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>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</div><div><br></div><div><a href="http://www.caam.rice.edu/~mk51/" target="_blank">https://www.cse.buffalo.edu/~knepley/</a><br></div></div></div></div></div>
</div></div>