AMBER Archive (2006)

Subject: AMBER: sequential execution of number of parallel jobs

From: In Hee Park (
Date: Mon May 01 2006 - 15:13:23 CDT

Dear Amber users:

I'd like to collect trajectories not in one file, but separately several
files as did in the AMBER Tutorial:

But queue job submi environment in order to use paralle sander, how to execute
(submit the queue jobs) a number of parallel jobs one by one in sequential manner,
i.e. after one parallel job is done, using its resulting file as input, then
execute the next job?

Could you let me know how to do that by which part should be modified in
the script below?

import sys, posix

path = "/path/dir/name"
amber = "/amber/dir/name/exe"

for i in range(2, 10+1, 1):
        j = i - 1
        jobname = """%s%d""" % ("a-dna_md", i)
        inpname = """%s%d""" % ("a-dna_md", j)
        number = """%s%d""" % ("MDINFO", i)
        command = """ echo "#PBS -N testSanderMPI
#PBS -l walltime=00:30:00
#PBS -l nodes=4:ppn=2:myrinet
#PBS -S /bin/bash
#PBS -j oe
cd %s
%s/sander.MPI -O -i -o %s.out -p a-dna_wat.prmtop -c %s.rst -r %s.rst -x %s.mdcrd -inf %s" > %s.script
chmod +x %s.script
qsub %s.script """ % (path, amber, jobname, inpname, jobname, jobname, number, jobname, jobname, jobname)


In Hee Park

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to