Convenience: SLURM distribution from a named profile

Builds a SLURM distribution spec using a resource profile name (or profile object) so you don't have to manually call slurm_resources() in common cases.

Usage

dist_slurm_profile(
  profile,
  by = NULL,
  within = c("multisession", "multicore", "callr", "sequential"),
  workers_within = NULL,
  template = slurm_template(),
  chunks_per_job = 1L,
  target_jobs = NULL
)

Arguments

profile

Profile identifier passed to slurm_resources(). Typically a character name registered via slurm_defaults_set() or profile_register(); can also be a parade_profile object or a plain list.

by

Optional column names to group by for parallelization.

within

Execution strategy within each SLURM job: "multisession", "multicore", "callr", or "sequential".

workers_within

Number of workers used within each SLURM job.

template

Path to SLURM template file.

chunks_per_job

Number of groups to process per SLURM job.

target_jobs

Optional integer; target number of SLURM jobs to create (overrides chunks_per_job at submit time).

Value

A parade_dist object suitable for distribute()

Examples

# Create SLURM distribution using a profile
if (FALSE) { # \dontrun{
dist <- dist_slurm_profile("standard", by = "group")
} # }

# Use multicore for within-job parallelism
if (FALSE) { # \dontrun{
dist <- dist_slurm_profile("highmem", within = "multicore", workers_within = 16)
} # }

# Configuration example (no execution)
if (FALSE) { # \dontrun{
flow(grid) |>
  stage("analyze", analyze_fn) |>
  distribute(dist_slurm_profile("standard"))
} # }