<div dir="ltr">Have you set the environment variable?<br>export OMP_NUM_THREADS=8<br></div><br><div class="gmail_quote"><div dir="ltr">On Wed, Jul 8, 2015 at 1:05 PM Ian Washington <<a href="mailto:washinid@mcmaster.ca">washinid@mcmaster.ca</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi All,<br>
<br>
Quick question.<br>
<br>
At what point in the problem size (variables/constraints) does using a<br>
parallel shared-memory linear solver like ma86, ma97 or pardiso become<br>
beneficial in terms of speedup.<br>
<br>
I am trying to do some timings on a sparse discretized optimal control<br>
problem with about 30000 variables and 20000 equalities (i.e. 10000<br>
dof). Currently, ma86 and pardiso require more time than the serial ma27<br>
solver for the same amount of iterations. I am using openmp 3.1 with<br>
gcc-4.7.3 on linux 64bit.<br>
<br>
I am guessing my problem is just too small at present.<br>
<br>
Anyone have any thoughts?<br>
<br>
Thanks,<br>
Ian.<br>
_______________________________________________<br>
Ipopt mailing list<br>
<a href="mailto:Ipopt@list.coin-or.org" target="_blank">Ipopt@list.coin-or.org</a><br>
<a href="http://list.coin-or.org/mailman/listinfo/ipopt" rel="noreferrer" target="_blank">http://list.coin-or.org/mailman/listinfo/ipopt</a><br>
</blockquote></div>