problem in writing scripts with multiple nodes

[sarkarde...@gmail.com]

Hello

I am trying to run a script. Suppose I have 16 jobs and I want to submit those jobs in 4 nodes, using 4 processor for each node. I am currently using the script

{
#PBS -o logfile.log
#PBS -e errorfile.err
#PBS -l select=4:ncpus=4:mpiprocs=4,walltime=1:0:0
#PBS -j oe
#PBS -N 1N8-fac
#PBS -q medium

pbsdsh -v $PBS_O_WORKDIR
#pbsdsh echo $PBS_VNODENUM

# DIRECTORY TO RUN - $PBS_O_WORKDIR is directory job was submitted from

echo $PBS_NODEFILE


cd $PBS_O_WORKDIR
PATH=$PBS_O_PATH

(cd run1 $PBS_VNODENUM; ./run1) &
(cd run2 $PBS_VNODENUM; ./run2) &
(cd run3 $PBS_VNODENUM; ./run3) &
(cd run4 $PBS_VNODENUM; ./run4) &
(cd run5 $PBS_VNODENUM; ./run5) &
(cd run6 $PBS_VNODENUM; ./run6) &
(cd run7 $PBS_VNODENUM; ./run7) &
(cd run8 $PBS_VNODENUM; ./run8) &
(cd run9 $PBS_VNODENUM; ./run9) &
(cd run10 $PBS_VNODENUM; ./run10) &
(cd run11 $PBS_VNODENUM; ./run11) &
(cd run12 $PBS_VNODENUM; ./run12) &
(cd run13 $PBS_VNODENUM; ./run13) &
(cd run14 $PBS_VNODENUM; ./run14) &
(cd run15 $PBS_VNODENUM; ./run15) &
(cd run16 $PBS_VNODENUM; ./run16) &

wait }


when I am submitting it , it is showing like the below

JobID Username Queue Jobname SessID NDS TSK Memory Time S Time

5222.kak.hpc. snig medium 1N8-fac 10673 4 16 -- 01:00 R 00:00 n5/0*4+n9/0*4+n10/0*4+n11/0*4

and the script is submitting the all 16 jobs in a single node with 16 processor, without dividing the jobs in multiple nodes. Is there any solution?? kindly give me some suggestion.

Thanks
Deabarti