- The program has been tested on vnfe4 cluster.
- The make file is set for the Portland compiler so use that one.
- If you plan to test it on the Gigabit nodes (node056-node057) set the
  environment by sourcing the included file 'soPGI-mpich':

source soPGI-mpich

- If you plan to run it through PBS on myrinet nodes
  then set your environment by executing:

source ~matt/scripts/soPGI-myr

- Compile with 'make'

- To test it interactively on 2 CPUs on the head node (can be compiled with either environment)
  execute:

mpirun -np 2 ./pamr_uni1d.exe

- To test it on the Gigabit nodes execute (for 4 CPUs) execute

mpirun -np 4 -machinefile machfile ./pamr_uni1d.exe

  where the machfile should contain the names of the nodes, e.g.,
node056
node057
node056
node057

- To submit to PBS use

qsub run.pbs

- By default the number of CPUs is set to 4.
  If you want to change it, change the 'nodes=4' in the line

#PBS -l nodes=4:ppn=1:myrinet

  of run.pbs to the desired number.
