Next: 4.2 Parallelization levels
Up: 4 Parallelism
Previous: 4 Parallelism
Contents
4.1 Running on parallel machines
Parallel execution is strongly system- and installation-dependent.
Typically one has to specify:
- a launcher program (not always needed),
such as poe, mpirun, mpiexec,
with the appropriate options (if any);
- the number of processors, typically as an option to the launcher
program, but in some cases to be specified after the name of the
program to be
executed;
- the program to be executed, with the proper path if needed: for
instance, ./neb.x, or $HOME/bin/neb.x, or
whatever applies;
- other PWscf-specific parallelization options, to be
read and interpreted by the running code;
- the number of image groups used by NEB (see next subsection).
Items 1) and 2) are machine- and installation-dependent, and may be
different for interactive and batch execution. Note that large
parallel machines are often configured so as to disallow interactive
execution: if in doubt, ask your system administrator.
Item 3) also depend on your specific configuration (shell, execution
path, etc).
Item 4) is optional but may be important: see the following section
for the meaning of the various options.
For illustration, here is how to run neb.x on 16 processors partitioned into
4 image groups (4 processors each), for a path containing at least 4 images with POE:
poe neb.x -procs 16 -ni 4 -i input
Next: 4.2 Parallelization levels
Up: 4 Parallelism
Previous: 4 Parallelism
Contents
Paolo Giannozzi
2017-10-23