[CSC 435] Class on Tuesday

[Andrew J. Pounds]

Guys -- with recent changes to encryption standards I found out today 
that I've got to rewrite and test my perl scripts that help you generate 
ssh keys for the cluster. I've got to think how to do this carefully so 
I don't impact my research calculations that are currently running on 
the cluster.  I won't get that done tonight.  As such tomorrow is going 
to be a "work day" to work on code.  I will be in the lab at 9:30 if you 
have any questions.  If you are not planning on coming to class please 
let me know.

Remember, you need to use the lab computers to test your code because 
they will give you consistent results (not NaN) for the trace and 
idcheck with MAXDIM=100.

Here are results for an early, non-optimal C program, the original 
Dusty, my modern Fortran dusty, and a modern Fortran dusty on which I 
did some major code rearrangement based on ideas from the handouts I 
gave you last week.  MAXDIM in all cases was 100.

gsc218-node-24:/tmp/dustydev % dusty;dusty77;dusty90;dusty08
Final trace = 3.867438128744303e+05 and IDCHECK 0.81989322403772058
  -- RUNTIME -> 22.669775 seconds
  Final trace =    386743.81287443032       and IDCHECK 
  0.81989322403772058
  -- RUNTIME ->    70.020658999999995       seconds
  Final trace =    386743.81287443032       and IDCHECK 
  0.81989322403772058
  -- RUNTIME ->    22.263003999999999       seconds
  Final trace =    386743.81287443032       and IDCHECK 
  0.81989322403772058
  -- RUNTIME ->    8.6740130000000004       seconds

-- 
*/Andrew J. Pounds, Ph.D./*
/Professor of Chemistry and Computer Science/
/Director of the Computational Science Program/
/Mercer University, Macon, GA 31207 (478) 301-5627/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20260119/3f541450/attachment.htm>