CD task8 protocol

From Bioinformatikpedia
Revision as of 09:56, 29 June 2012 by Reeb (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Preparation

  • Load LRZ tools:
module load lrztools
  • Load git
module load git
  • clone git repository:
git clone git://github.com/offmarc/AGroS.git
  • SetPath
export PATH=$PATH:$HOME/AGroS/
export PATH=$PATH:$HOME/software/scwrl/
rename Scwrl4 to scwrl
  • Job script for testing: job.sh
#!/bin/bash
#SBATCH -o /home/hpc/pr58ni/di34fog/jobs/wt/wt_md.out
#SBATCH -D /home/hpc/pr58ni/di34fog/jobs/wt
#SBATCH -J wt_MD
#SBATCH --partition=mpp1_inter
#SBATCH --get-user-env
#SBATCH --ntasks=32
#SBATCH --mail-type=end
#SBATCH --mail-user=susann.vorberg@cmapus.lmu.de
#SBATCH --export=NONE
#SBATCH --time=02:00:00
source /etc/profile.d/modules.sh
module load gromacs
export PATH="$HOME/AGroS:$PATH"
export PATH="$HOME/apps/bin/:$PATH"
export PATH="$HOME/software/scwrl/:$PATH"
AGroS 2O4H_chainA.pdb --scwrlPATH /home/hpc/pr58ni/di34fog/software/scwrl -dir /home/hpc/pr58ni/di34fog/jobs/wt -threads 32

for real run:

--time 32:00:00
--partitition=mpp1
    • Watch out, --partition is not the same as --clusters. Maybe check the sbatch manual again. I'm not completely sure what might happen either, but since you will probably also be waiting quite some time until your job finally starts, maybe you shouldn't take any chances and change it to --clusters=mpp1? -jonas


  • Log in and submit job
ssh -X lxia4-1
sbatch job.sh
  • Status:
squeue -u di34fog --cluster mpp1
squeue --cluster mpp1