SLURM resources as ‘partition’, ‘ntasks’, ‘cpus’, etc. need to be defined per rule

[pmb59]

After the SLURM migration, we can fine tune even more the resources for each rule:

Execute snakemake rules as SLURM batch jobs according to their ‘resources’ definition. SLURM resources as ‘partition’, ‘ntasks’, ‘cpus’, etc. need to be defined per rule within the ‘resources’ definition. Note, that memory can only be defined as ‘mem_mb’ or ‘mem_mb_per_cpu’ as analogous to the SLURM ‘mem’ and ‘mem-per-cpu’ flags to sbatch, respectively. Here, the unit is always ‘MiB’. In addition ‘–default_resources’ should contain the SLURM account.

https://snakemake.readthedocs.io/en/stable/executing/cli.html#slurm