Terry Frankcombe a écrit :
Thanks for your suggestions.
I'm no SGE expert. But don't you have a PE available that simply
allocates nodes and calls your script? Then you can specify in your
script any mpirun you want, and it all should still work.
Alternatively, can't you shut down the SGE-called mpirun as the first
thing you do, then continue on calling your own mpirun? All this
depends on exactly how your machine and SGE is set up.
I don't see how statically linking your app avoids this issue at all, if
you're still calling the wrong mpirun.
My trials are not a prioritary task in my duty, and furthermore not
really done in a scientific way I must admit, so that I could very well
have misunderstood what was the signification of the problems that
prevented me from executing my code using faster
settings/compilers/libraries than the default one. (I have made quite a
lot of attempts but each time during a small amount of time).
The important thing for me now is that I can at last develop on my own
new PC and test my code there (the dynamic code stops before doing
anything when run). I will reconsider things, when we will have a new
cluster around october.
Sorry for the confusion in my messages.