A package to download slurm usage from a supercomputer and upload it to MongoDB
Slurm-Mongo is a Python package that downloads SLURM usage data from a supercomputer and uploads it to MongoDB.
You can install Slurm-Mongo using pip:
pip install slurm-mongo
You can use Slurm-Mongo by running the slurm_mongo
command followed by the start date, end date, and supercomputer host name in user@hostname format:
slurm_mongo start_date end_date supercomputer_host_name
slurm_mongo '2017-01-01' '2017-01-02' 'shinde@snellius.surf.nl'
By default, Slurm-Mongo will connect to a MongoDB instance running on localhost:27017
and will store the data in a database called "Snellius" and a collection called "Usage". You can customize the MongoDB connection and database/collection names by using the optional --connection
, --port
, --database
, and --collection
flags:
slurm_mongo start_date end_date supercomputer_host_name --connection mongo.example.com --port 12345 --database MySlurm --collection MyUsage
Slurm-Mongo is released under the MIT License. See LICENSE for more information.
If you encounter any issues with Slurm-Mongo, please report them on the GitHub issue tracker.
Contact the author Dr. Ravindra Shinde at r.l.shinde@utwente.nl