Node dependent files, with a copy per node
Each node creates a file with a basename which is the same as in the
serial version, with an additional suffix. The suffix is a string
which shows the node ID. All
files are written into a temporary directory and the
processor ID is the most convenient way for referring to each file.
Note, that this method ensures that both applications,
nonh_mpi and mchf_mpi refer to precisely the same file
as the node ID. This is important for using parallel
environments which do not explicitly specify the processor nodes.
-rw-r--r-- 1 georgio georgio 6619380 Sep 20 00:35 c.lst.000
-rw-r--r-- 1 georgio georgio 863484 Sep 20 00:35 ico.lst.000
-rw-r--r-- 1 georgio georgio 393736 Sep 20 00:35 ih.lst.000
-rw-r--r-- 1 georgio georgio 16760 Sep 20 00:35 yint.lst.000