I am using this script running MD simulation on AMBER.
#!/bin/bash
#PBS -N eric.job
#PBS -l select=1:ncpus=12:mpiprocs=12
#PBS -l walltime=1:00:00
#PBS -q kepla_k20
#PBS -m be
#PBS -o /export/home/kbisetty/scratch5/amber/faez/test/tion.out
#PBS -e /export/home/kbisetty/scratch5/amber/faez/test/tion.err
#PBS -M [email protected]
cd $PBS_O_WORKDIR
pwd
echo "My job starts here"
date
source /etc/profile.d/modules.sh
module add intel-XE/c8000 cuda/5.0-c8000 mvapich2/c8000 amber/k20
module add amber/k20
exe=pmemd.cuda.MPI
nproc=`cat $PBS_NODEFILE|wc -l`
time mpirun -np $nproc -machinefile $PBS_NODEFILE $exe -O -i md.in -o md.out -p prmtop -c inpcrd
echo "My job ends here"
date
md.in file:
Initial minimization of solvent and solute, 10.0 cut
&cntrl
imin = 1,
ntb = 1, ntc = 2, ntf =2,
cut = 10.0,
maxcyc = 500,
ncyc = 500,
/
PLease let ,me know any error here. I am generating an error output files as:
PMEMD implementation of SANDER, Release 12
| Run on 07/15/2014 at 04:00:21
[-O]verwriting output
File Assignments:
| MDIN: md.in
| MDOUT: md.out
| INPCRD: inpcrd
| PARM: prmtop
| RESTRT: restrt
| REFC: refc
| MDVEL: mdvel
| MDEN: mden
| MDCRD: mdcrd
| MDINFO: mdinfo
|LOGFILE: logfile
Here is the input file:
Initial minimization of solvent and solute, 10.0 cut
&cntrl
imin = 1,
ntb = 1, ntc = 2, ntf =2,
cut = 10.0,
maxcyc = 500,
ncyc = 500,
/
CUDA (GPU): Minimization is NOT supported in parallel on GPUs. Please use
the single GPU code for minimizations.
Input errors occurred. Terminating execution.
"md.out" 42L, 2132C
I would suggest that you perform minimization, heating and equilibration of your system in CPU version of amber ( for example, using pmemd.MPI) and do the production dynamics in cuda version of amber (pmemd.CUDA). If you have multiple GPU you can use parallel version of CUDA for running MD on multiple GPU. I hope it helps.
Dear, I believe if you compile it carefully again will work. For the error in the folder'' bin / bash'' is the installation directory
here you can find the options that are supported in pmemd http://ambermd.org/gpus/
I would suggest that you perform minimization, heating and equilibration of your system in CPU version of amber ( for example, using pmemd.MPI) and do the production dynamics in cuda version of amber (pmemd.CUDA). If you have multiple GPU you can use parallel version of CUDA for running MD on multiple GPU. I hope it helps.
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry
- Computational Chemistry
- Molecular Dynamics