<div dir="ltr">This sounds like a hypre issues. <span style="font-family:arial,sans-serif;font-size:13px">Euclid/Hypre is not use much AFAIK so it might have some problems that have not been addressed.</span></div><div class="gmail_extra">
<br><br><div class="gmail_quote">On Tue, Nov 19, 2013 at 12:11 PM, Manav Bhatia <span dir="ltr"><<a href="mailto:bhatiamanav@gmail.com" target="_blank">bhatiamanav@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>Hi, </div>
<div> </div>
<div> I am using ilu(k) preconditioner from Euclid/Hypre on my Mac OS 10.9 box with 12 processors and openmpi v1.7.3 from macports. </div>
<div> </div>
<div> I find that for 1 CPU (-np 1) the code runs reasonably fast, but when multiple processors are included the solver runs significantly slow (by orders of magnitude). I ran the time profiler on one of the processes spawned by openmpirun, and it seems to be taking a lot of time in MPI_Waitall called from Euclid_dhSetup. </div>
<div> </div>
<div> This could very well be a Hypre issue, but I wanted to check to see if anynoe on the Petsc list has experience a similar problem. </div>
<div> </div>
<div> Any suggestions/inputs are welcome. </div>
<div> </div>
<div>Thanks,</div>
<div>Manav</div>
<div> </div>
</blockquote></div><br></div>