The main purpose of this wiki is to enable comfortable editing of the skill definitions from the git repository and for the available certificates. One skill or certificate is exactly one page.
Note that the chairs for the subtree may comment or discuss the proposal.
A skill is located in a unique place in the skill tree, which is the URI. There may be multiple references to a skill in the tree (which is then like a graph).
A skill is defined by the following fields:
This skill is made up for demonstration purposes. Firstly, we have a high-level skill.
Now, let's see a generic sub-skill, this extends the previous definition being a sub-skill. Hence, it is intended that the objectives and outcomes from the parent skill above are covered and refined into specifics.ID: USE4.2.1-BName: Workload manager introductionBackground
Outcomescomprehend the exclusive and shared usage model in HPCdifferentiate batch and interactive job submissioncomprehend the generic concepts and architecture of resource manager, scheduler, job and job scriptexplain environment variables as a means to communicatecomprehend accounting principlesexplain the generic steps to run and monitor a single job
Now, let's see a specific sub-skill for a specific software, this also extends the previous definition being a sub-skill but is on the same level as the basic skill above. Hence, multiple similar software solutions can build on the knowledge disseminated in USE4.2.1-B.ID: USE4.2.2-BName: SLURM Workload managerBackground
Outcomesrun interactive jobs with salloc, a batch job with sbatchexplain the architecture of SLURM, i.e., the role of slurmd, srun and the injection of environment variablesexplain the function of the tools: sacct, sbatch, salloc, srun, scancel, squeue, sinfoexplain time limits and the benefit of a backfill schedulercomprehend that environment variables are set when running a jobcomprehend and describe the expected behavior of a simple job scriptscomprehend how variables are prioritized when using command line and a scriptchange a provided job template and embed them into shell scripts to run a variety of parallel applicationsanalyze the output generated from submitting to the job scheduler and typically generated errors