This repository stores some useful pipelines used in Pawsey.
This script is a wrapper generating a job submission script to run multiple jobs in magnus, zeus and raijin. It automatically decides the number of array and jobsPerNode to fully use the resources. Version: 0.83 USAGE: ./runJobs.sh [-h] [-d] [-c [<configFileName>]] [-C <magnus|zeus|raijin>] [-q <nodeType>] [-a <STR>] [-i <INT>] [-V] <configFile> -h This help. -d Debug mode, only generate the submission script but not submit the job. -c Generate a template configFile with the optional config filename. Default name: thisJob.conf. Advanced options: -C Cluster to be run on <magnus|zeus|raijin>. Default current server, eg. magnus -q Node type to be submit to, default magnus:workq; zeus:workq; raijin:normal -a Set the account to be used. Default pawsey0149. Options for pawsey servers: -i Interval of memory probing in job, unit in seconds. Default 30. -V No verbose - disable memory usage checking for jobs.
Description: This script is a wrapper to wrap around the sbatch job submission. Automatically do batch submission based on the system config limit. USAGE: bash ./submitArray.sh <jobSubmitssionOptions> <jobScript> example: bash ./submitArray.sh --array=1-300 mySlurmScript.slm