I have just finished a new version of a parallel code that I have been working on. I would like to do a strong scaling test, but I am always bothered by how much monotonous work scaling tests seem to require.
Currently, I make folders for each processor count (np1, np2, np4, ...), copy my code into each folder, assign the correct parameters to the PBS file for submission to the cluster, and then submit the jobs one-by-one.
After job completion, I copy the data from each folder onto my machine and peice-by-piece collect the timing data from each folder. Then it goes into MATLAB.
I am curious if any of you have developed more efficient ways to run scaling tests. Do you write scripts to do most of this for you? If so, how do you accomplish that? Bash scripting? Perl, perhaps?