<div class="gmail_quote">On Tue, Nov 29, 2011 at 23:53, Dave Nystrom <span dir="ltr"><<a href="mailto:dnystrom1@comcast.net">dnystrom1@comcast.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div id=":4pv">I have a resistive mhd code that I have recently interfaced to petsc which<br>
has 7 linear solves that are all symmetric. I recently tried using -pc_type<br>
cholesky -ksp_type preonly for a run and found that it was taking about 6<br>
times as long per linear solve as when I was using -pc_type lu -ksp_type<br>
preonly. </div></blockquote><div><br></div><div>Try -pc_factor_mat_ordering_type nd</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div id=":4pv"> I was wondering if that was reasonable behavior. I would not have<br>
thought that using a cholesky direct solve would take longer than an LU<br>
direct solve in petsc for the serial case and was hoping it would be faster.<br>
Does this behavior seem reasonable?</div></blockquote></div><div><br></div>Try this:<div><br><div>-pc_type cholesky -pc_factor_mat_ordering_type nd</div></div><div><br></div><div>Barry, why is natural ordering still the default for Cholesky? It is so slow that it is worthless.</div>