Each directory contains an example usage of the simple slurm profile. You can run them on your cluster to confirm the expected behavior.
-
cluster-cancel
- Cancel all submitted jobs withscancel
when the main Snakemake process is cancelled (requires Snakemake 7.0.0+) -
conditional-resource
- Conditionally specify a resource for a subset of rules -
dynamic-resources
- Increase memory for a rule on restart after each failed attempt -
job-array
- Use a Slurm job array to quickly execute a rule that must be run thousands of times -
job-grouping
- Adapt profile to allow job grouping -
jobs-per-second
- Measure how many jobs Snakemake can submit to Slurm per second -
list-partitions
- List multiple partitions for Slurm to choose from -
log-file-date
- Insert the current date into the log file path -
many-input-files
- Use a custom jobscript when a rule has many thousands of input files -
mem-gb
- Specify memory in GB -
multi-cluster
- Submit jobs to more than one cluster (requires Snakemake 7.1.1+) -
out-of-memory
- Trigger an out-of-memory error. Snakemake can handle this by default -
pinned-conda-envs
- Pin packages in conda environments -
separate-stderr
- Log stdout and stderr to separate files -
shared-logs
- Share the same profileconfig.yaml
between multiple pipelines, but include the name of the Snakefile in the path to the log files -
singularity
- Run a rule inside a singularity container -
slurm-native
- Use Snakemake's native Slurm support (--slurm
, requires Snakemake 7.19.0+) -
threads
- Increase thethreads
for a rule. It is responsive to--jobs
. In other words, it will scale down the number of threads used if there aren't enough available (thus make sure to pass{threads}
to the software you are running). This mainly applies when running the rules locally (i.e. without the profile). When submitting to the cluster,--jobs
refers literally to jobs, not cores like in local mode -
time-integer
- Specify--time
as an integer (minutes) -
time-string
- Specify the time as"days-hours:minutes:seconds"
-
timeout
- Trigger a timeout error. Snakemake requires the assistance of a custom script to--cluster-status
to handle this error. Otherwise it just stalls indefinitely