[CSC 435] Today's Class and MPI

[Andrew J. Pounds]

I spent part of Monday night, yesterday, and this morning working on 
converting my distributed-parallel MPI exercise to work on the Godsey 
cluster.    The link to that exercise is

https://docs.google.com/document/d/1cIKDqVY6bKfFYqiPmlid_vlJ1RQO9N5v4-JX3CmP6lw/edit?usp=sharing

One thing I did learn is that if more than one of us does this 
simultaneously it will be a VERY painful exercise ( we will fix that 
with when we use the batch scheduler).  So, instead of us running the 
codes as a class this afternoon, during our ZOOM time at 3 pm I will 
walk you through the MPI programs that you will be running.  I 
anticipate today's lecture to be pretty short.  I will, however, need 
you to try and work through the exercise after we meet today and tell me 
where you had problems.   We MUST get this stuff sorted out before we 
wade into bigger MPI codes.

While Friday is "Good Friday" and there are no classes, computers don't 
have souls so I encourage you to let them work on running code while you 
take the day off.  I'm still putting together the MPI assignment that 
deals with Jacobi tiling.  My experience over the last 24 hours with 
Godsey cluster has taught me a lot about what is possible and I had to 
make some revisions.

I will have the ZOOM link to you for today's class in a little while.

-- 
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