[CSC 435] RULES OF ENGAGEMENT

Andrew J. Pounds pounds_aj at mercer.edu
Mon Apr 30 17:54:05 EDT 2018


There has been some confusion about what we are doing tomorrow.

During the final tomorrow all lab systems will be shutdown.  Nobody may
work on code once the final begins.

After the final EACH GROUP will

1. Come to the front of the class and clone your repo into a directory I
specify on hammer.

2. Describe what optimization techniques you applied to your C and
Fortran code.  A brief PowerPoint or Google Sheet would be awesome.

3. RUN both codes on a maximum of 40 threads using hammer.  Note - you
do not have to use 40 threads.  40 threads is the maximum you can use.


I will specify the dimension of the of the box for each run.  Fastest
average runtime to equilibrium wins.


BECAUSE HAMMER IS BEING USED SO MUCH RIGHT NOW FOR OPENBLAS TIMING, YOU
DO NOT HAVE TO BUILD PERFORMANCE GRAPHS.


Tomorrows results should be considered preliminary.  I will be testing
your codes on another system (similar architecture to hammer) that is
"quiet" so we get accurate test results. 


-- 
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/20180430/2a9faf83/attachment.html>


More information about the csc435 mailing list