Tools to keep track of jobs submitted using PBS torque.

Often, users of an HPC environment submit many jobs from a multitude of directories, and it is not always trivial to do the bookkeeping.

jobdb will help with that by recording the job ID, the submission time, the directory it was submitted in, and, optionally, the job name if specified on the command line with the -N option, and a description with the --description option.

It is now easy to navigate to the directory using jdbcd for a specified job ID, or choose it interactively from a list that can be filtered.

• jdbsub: qsub wrapper that keeps track of the job's meta-information.
• jdbcd: change directory to a job's submission directory.
• jdblist: show the jobdb file in your favorite pager (defaults to less).
• jdbedit: edit the jobdb file in your favorite editor.
• jdbclear: clear all entries in the jobdb database file.

All commands will provide a help message when the --help option is used. Documentation for peco can be found on its GitHub site.

Simply source the jobdb.sh file from your .bashrc file, i.e.,

source "${HOME}/jobdb.sh"

peco should be in your path. It can be downloaded from GitHub. Pre-built binaries work just fine.

jdbsub will update a job database file jobs that is by default located in the .jobdb directory in your home directory. Its location can be controlled using the JOBDB_DIR environment variable if required.

The "database" is a CSV file that keeps track of the job ID as returned by qsub, the job name if specified on the command line using qsub's -N option, the directory in which jdbsub was executed, a description if provided using the --description option, and a timestamp in the format YYYY-mm-dd HH:MM:SS.