<html><head><meta http-equiv="Content-Type" content="text/html charset=iso-8859-1"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div>As Jed mentioned you should not see a significant difference with '-random_seed 1'. This fix may just be papering over a problem that will bite you eventually. If you run with '-XXX_gamg_verbose 2' the eigenvalues used will be printed out. Try running with and without '-random_seed 1' and send the output. We are looking to see if the computed eigenvalues on these second (bad) solves is different in the two runs. </div><div><br></div><div>And I'm a bit puzzled because I don't know how '-random_seed 1' could be used in GAMG to compute eigen estimates...</div><div><div><br></div><div><br><div><div>On Jul 11, 2013, at 4:47 PM, John Mousel <<a href="mailto:john.mousel@gmail.com">john.mousel@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div dir="ltr">I have two KSP contexts, helm%ksp and proj%ksp. I switch between the two by calling KSPSetOperators. There was no performance difference, just a different answer. <br></div><div class="gmail_extra"><br><br><div class="gmail_quote">
On Thu, Jul 11, 2013 at 3:41 PM, Jed Brown <span dir="ltr"><<a href="mailto:jedbrown@mcs.anl.gov" target="_blank">jedbrown@mcs.anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0px 0px 0px 0.8ex; border-left-width: 1px; border-left-color: rgb(204, 204, 204); border-left-style: solid; padding-left: 1ex; position: static; z-index: auto; ">
<div class="im">John Mousel <<a href="mailto:john.mousel@gmail.com">john.mousel@gmail.com</a>> writes:<br>
<br>
> Jed,<br>
><br>
> I alternate between solving Helmholtz and Poisson equations in an outer<br>
> loop. The matrix is fixed for both operators during the process, and only<br>
> the right-hand side is updated for each system.<br>
<br>
</div>Do you have one KSP or two? If one, I assume you're trying to preserve<br>
memory, but there is likely more memory in the preconditioner than the<br>
Krylov space and if you want to reuse the preconditioner, you have to<br>
set it up anew each time.<br>
<div class="im"><br>
> I just tried -random_seed 1, and it seems to have made all the difference<br>
> in the world!<br>
<br>
</div>That's disconcerting because we would like the algorithm to be<br>
insensitive to such things. How big was the performance difference?<br>
Can you give us some information to reproduce? (Maybe a smallish<br>
example matrix that demonstrates this problem.)<br>
</blockquote></div><br></div>
</blockquote></div><br></div></div></body></html>