I am creating a model with repast simphony and I am trying to conduct batch runs on a cluster. More specifically, I am using slurm. I tested my batch run configuration on my local machine, and my results were exported and aggregated as expected.
However, when I use the model archive (.jar file) and run it on a cluster using slurm, then my model only runs once, rather than multiple times. I know this because when I check my outputs after running the ./outputcombiner.sh command, I only have results from one run.
My process includes unzipping the complete_model.jar file in the cluster, then editing the repast.slurm file, checking the batch_params.xml file (that says runs=X), then running the *chmod +x .sh command, and finally running the sbatch repast.slurm. **Is there a step I am missing when setting up the batch runs on the cluster? **
I tried many methods to solve this issue. I tried editing the batch_params.xml file and the repastwrapper.sh file to ensure that runs was greater than 1. This still resulted in my model only running once. I tried changing the parameters in the repast.slurm file, but for now, I have kept--
`#!/bin/bash
#SBATCH --job-name=MYJOB
#SBATCH --partition=normal
#SBATCH --nodes=1
#SBATCH --ntasks-per-node=1
#SBATCH --time=0-02:00:00
#SBATCH --output=output.txt
#SBATCH --mem-per-cpu=2GB
module load openjdk `
