SLURM - job scheduler that managers the queue and determines the order in which jobs are executed

Youtube video onboarding -

https://youtu.be/Pak8OOLDNkQ

Onboarding Steps:

1. Join #computing on slack
2. request a FASRC account https://portal.rc.fas.harvard.edu/request/account/new for sneel lab
3. Join our Github : https://github.com/safr-ai-lab/Cluster-Utils-Files/tree/main/FASRC

Resources for using FASRC

• Harvard FASRC Cluster Tips: https://docs.google.com/document/d/1-cGIzz30biYh-3FsFzaq0vQoXeXLY3ZWaFGkayAznq8/edit?pli=1#heading=h.ao9iifmp307h
• FASRC running jobs webpage https://docs.rc.fas.harvard.edu/kb/running-jobs/
• Kempner cluster onboard (David B.’s version): https://docs.google.com/document/d/12xqwQ9L4BU-WTWgB0JZ4WiUL5zndXduSKIEB9Dh8mik/edit#heading=h.gh8gp3ootk4e

How to install python libraries

never run anything serious on the login node, (to set things up, run something interactively)

Like this - CPU only

    #!/bin/bash
    salloc -p test --mem=20G -t 0-06:00

with gpu:

    #!/bin/bash
    salloc -p gpu_test --mem=20G -t 0-06:00

Now set up a python virtual environment (if you don’t know what this, please ask someone)