[CSC 435] Good Progress

[Andrew J. Pounds]

It looks like several of you, based on gitlab check-ins, now have code 
that is consistent across your languages.  I would encourage you to 
start your "no optimization" timings using the batch scheduler as soon 
as possible - some of these things will take over an hour to run.

My recommendation is that you modify your pbs job script so that the 
line that looks like

#PBS -l nodes=1:zeus:ppn=1

   is modified to

#PBS -l nodes=1:zeus:ppn=1,walltime=8:00:00

This will then give you eight hours of run time.  Adjust the time as you 
see fit.

I also recommend that you do at least three timings per code.   I have 
seen the runtimes for very long running calculations vary by minutes.   
You will want to average to get good results.

In other words, in your job submission do something like..

#.end of PBS commands
#
cd ./mytimingdirectory
echo "*** F77 ***"
dusty_deck_77
dusty_deck_77
dusty_deck_77

I'll see you in the queue!

-- 
Andrew J. Pounds, Ph.D.  (pounds_aj at mercer.edu)
Professor of Chemistry and Computer Science
Mercer University,  Macon, GA 31207   (478) 301-5627
http://faculty.mercer.edu/pounds_aj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20160222/33964f63/attachment.html>