batchtools_slurm: A batchtools Slurm backend resolves futures in parallel via a...

In future.batchtools: A Future API for Parallel and Distributed Processing using 'batchtools'

A batchtools Slurm backend resolves futures in parallel via a Slurm job scheduler

Description

A batchtools Slurm backend resolves futures in parallel via a Slurm job scheduler

Usage

batchtools_slurm(
  ...,
  template = "slurm",
  scheduler.latency = 1,
  fs.latency = 65,
  resources = list(),
  delete = getOption("future.batchtools.delete", "on-success"),
  workers = getOption("future.batchtools.workers", default = 100L)
)

Arguments

template	(optional) Name of job-script template to be searched for by batchtools::findTemplateFile(). If not found, it defaults to the templates/slurm.tmpl part of this package (see below).
scheduler.latency	[numeric(1)] Time to sleep after important interactions with the scheduler to ensure a sane state. Currently only triggered after calling submitJobs.
fs.latency	[numeric(1)] Expected maximum latency of the file system, in seconds. Set to a positive number for network file systems like NFS which enables more robust (but also more expensive) mechanisms to access files and directories. Usually safe to set to 0 to disable the heuristic, e.g. if you are working on a local file system.
resources	(optional) A named list passed to the batchtools job-script template as variable resources. This is based on how batchtools::submitJobs() works, with the exception for specially reserved names defined by the future.batchtools package; resources[["details"]], if TRUE, results in the job script outputting job details and job summaries at the beginning and at the end. resources[["startup"]] and resources[["shutdown"]] are character vectors of shell code to be injected to the job script as-is. resources[["modules"]] is character vector of Linux environment modules to be loaded. resources[["envs"]], is an optional names character vector specifying environment variables to be set. resources[["rscript"]] is an optional character vector specifying how the 'Rscript' is launched. The resources[["rscript_args"]] field is an optional character vector specifying the 'Rscript' command-line arguments. resources[["asis"]] is a character vector that are passed as-is to the job script and are injected as job resource declarations. All remaining resources named elements are injected as named resource specification for the scheduler.
delete	Controls if and when the batchtools job registry folder is deleted. If "on-success" (default), it is deleted if the future was resolved successfully and the expression did not produce an error. If "never", then it is never deleted. If "always", then it is always deleted.
workers	The maximum number of workers the batchtools backend may use at any time, which for HPC schedulers corresponds to the maximum number of queued jobs. The default is getOption("future.batchtools.workers", 100).
...	Not used.

Details

Batchtools Slurm futures use batchtools cluster functions created by makeClusterFunctionsSlurm2(), which are used to interact with the Slurm job scheduler. This requires that Slurm commands sbatch, squeue, sacct, and scancel are available on the current machine.

The default template script templates/slurm.tmpl can be found in:

system.file("templates", "slurm.tmpl", package = "future.batchtools")

and comprises:

#!/bin/bash
######################################################################
# A batchtools launch script template for Slurm
#
# Author: Henrik Bengtsson 
######################################################################

## Job name
#SBATCH --job-name=<%= job.name %>
## Direct streams to logfile
#SBATCH --output=<%= log.file %>

## Resources needed
<%
  ## Should scheduler "details" be seen?
  details <- isTRUE(resources[["details"]])
  resources[["details"]] <- NULL

  ## Shell "startup" code to evaluate
  startup <- resources[["startup"]]
  resources[["startup"]] <- NULL
     
  ## Shell "shutdown" code to evaluate
  shutdown <- resources[["shutdown"]]
  resources[["shutdown"]] <- NULL
     
  ## Environment modules specifications
  modules <- resources[["modules"]]
  resources[["modules"]] <- NULL
  
  ## Environment variables to be set
  envs <- resources[["envs"]]
  if (length(envs) > 0) {
    stopifnot(is.character(envs), !is.null(names(envs)))
  }
  resources[["envs"]] <- NULL
     
  ## Custom "Rscript" command and Rscript arguments
  rscript <- resources[["rscript"]]
  if (is.null(rscript)) {
    rscript <- "Rscript"
  } else if (length(rscript) == 0 || !nzchar(rscript)[1]) {
    stop("Argument 'resources' specifies an empty 'rscript' field")
  }
  resources[["rscript"]] <- NULL
  rscript_args <- resources[["rscript_args"]]
  resources[["rscript_args"]] <- NULL
  rscript_call <- paste(c(rscript, rscript_args), collapse = " ")
    
  ## As-is resource specifications
  job_declarations <- resources[["asis"]]
  resources[["asis"]] <- NULL
    
  ## Remaining resources are assumed to be of type '<key>=<value>'
  opts <- unlist(resources, use.names = TRUE)
  opts <- sprintf("%s=%s", names(opts), opts)
  job_declarations <- sprintf("#SBATCH %s", c(job_declarations, sprintf("--%s", opts)))
  writeLines(job_declarations)
%>

## Bash settings
set -e          # exit on error
set -u          # error on unset variables
set -o pipefail # fail a pipeline if any command fails
trap 'echo "ERROR: future.batchtools job script failed on line $LINENO" >&2; exit 1' ERR

<% if (length(job_declarations) > 0) {
  writeLines(c(
    "echo 'Job submission declarations:'",
    sprintf("echo '%s'", job_declarations),
    "echo"
  ))
} %>

<% if (details) { %>
if command -v scontrol > /dev/null; then
  echo "Job information:"
  scontrol show job "${SLURM_JOB_ID}"
  echo
fi
<% } %>

<% if (length(startup) > 0) {
  writeLines(startup)
} %>

<% if (length(modules) > 0) {
  writeLines(c(
    "echo 'Load environment modules:'",
    sprintf("echo '- modules: %s'", paste(modules, collapse = ", ")),
    sprintf("module load %s", paste(modules, collapse = " ")),
    "module list"
  ))
} %>

<% if (length(envs) > 0) {
  writeLines(c(
    sprintf("echo 'Setting environment variables: [n=%d]'", length(envs)),
    sprintf("echo ' - %s=%s'", names(envs), shQuote(envs)),
    sprintf("export %s=%s", names(envs), shQuote(envs))
  ))
} %>

echo "Session information:"
echo "- timestamp: $(date +"%Y-%m-%d %H:%M:%S%z")"
echo "- hostname: $(hostname)"
echo "- Rscript call: <%= rscript_call %>"
if ! command -v <%= rscript[1] %> &> /dev/null; then
    >&2 echo "ERROR: Argument 'resources' specifies a non-existing 'Rscript' launch command: <%= rscript[1] %>. Maybe you need to specify which environment modules to load in the 'resources' argument, e.g. 'plan(future.batchtools::batchtools_slurm, resources = list(modules = c(\"r\")))'. The search PATH for '<%= rscript[1] %>' was ${PATH}"
    exit 1
fi
echo "- Rscript version: $(<%= paste(rscript, collapse = " ") %> --version)"
echo "- R_LIBS_USER=${R_LIBS_USER:-<not set>}"
echo "- R_LIBS_SITE=${R_LIBS_SITE:-<not set>}"
echo "- R_LIBS=${R_LIBS:-<not set>}"
echo "- Rscript library paths: $(<%= rscript_call %> -e "cat(shQuote(.libPaths()), sep = ' ')")"
echo

## Launch R and evaluate the batchtools R job
echo "Calling 'batchtools::doJobCollection()' ..."
echo "- job name: '<%= job.name %>'"
echo "- job log file: '<%= log.file %>'"
echo "- job uri: '<%= uri %>'"
<%= rscript_call %> -e 'batchtools::doJobCollection("<%= uri %>")'
res=$?
echo " - exit code: ${res}"
echo "Calling 'batchtools::doJobCollection()' ... done"
echo

<% if (details) { %>
if command -v sstat > /dev/null; then
  echo "Job summary:"
  sstat --format="JobID,AveCPU,MaxRSS,MaxPages,MaxDiskRead,MaxDiskWrite" --allsteps --jobs="${SLURM_JOB_ID}"
fi
<% } %>

<% if (length(shutdown) > 0) {
  writeLines(shutdown)
} %>

echo "End time: $(date +"%Y-%m-%d %H:%M:%S%z")"

## Relay the exit code from Rscript
exit "${res}"

This template and the built-in makeClusterFunctionsSlurm2() have been verified to work on a few different Slurm HPC clusters;

1. Slurm 21.08.4, Rocky Linux 8, NFS global filesystem, UCSF C4 HPC environment (June 2026)

2. Slurm 22.05.10, Rocky Linux 9, Lustre global filesystem (September 2025)

3. Slurm 22.05.11, Rocky Linux 8, NFS global filesystem (September 2025)

4. Slurm 23.11.4, Ubuntu 24.04 LTS, NFS global filesystem, UC Berkeley SCF HPC environment (June 2026)

5. Slurm 24.11.3, AlmaLinux 9, Lustre global filesystem (September 2025)*

6. Slurm 24.11.5, Rocky Linux 9, VAST global filesystem, UCSF CoreHPC environment (June 2026)

(*) Verified with future.batchtools 0.20.0, which used batchtools::makeClusterFunctionsSlurm() since superseeded by makeClusterFunctionsSlurm2().

Value

Nothing.

References

• https://en.wikipedia.org/wiki/Slurm_Workload_Manager

Examples

library(future)

# Limit runtime to 10 minutes and memory to 400 MiB per future,
# request a parallel environment with four slots on a single host.
# Submit to the 'freecycle' partition. Load environment modules 'r' and
# 'jags'. Report on job details at startup and at the end of the job.
plan(future.batchtools::batchtools_slurm, resources = list(
  time = "00:10:00", mem = "400M",
  nodes = 1, ntasks = 4, "cpus-per-task" = 1,
  partition = "freecycle",
  modules = c("r", "jags"),
  details = TRUE
))

f <- future({
  data.frame(
     hostname = Sys.info()[["nodename"]],
           os = Sys.info()[["sysname"]],
    osVersion = utils::osVersion,
        cores = unname(parallelly::availableCores())
  )
})
info <- value(f)
print(info)

# As above, but use R from the Rocker 'r-base' Linux container;
#
#   mkdir -p ~/lxc
#   apptainer build ~/lxc/rocker_r-base.sif docker://rocker/r-base
#
# Example assumes that 'future.batchtools' has already been installed in
# the container to the 'R_LIBS_USER' package folder living on the host;
#
#   R_LIBS_USER=~/R/rocker-%p-library/%v ~/lxc/rocker_r-base.sif
#   ...
#   > chooseCRANmirror(ind = 1)
#   > install.packages("future.batchtools")
#
plan(future.batchtools::batchtools_slurm, resources = list(
  time = "00:10:00", mem = "400M",
  nodes = 1, ntasks = 4, "cpus-per-task" = 1,
  partition = "freecycle",
  details = TRUE,
  envs = c(R_LIBS_USER = "~/R/rocker-%p-library/%v"),
  rscript = c("apptainer", "exec", "~/lxc/rocker_r-base.sif", "Rscript")
))

f <- future({
  data.frame(
     hostname = Sys.info()[["nodename"]],
           os = Sys.info()[["sysname"]],
    osVersion = utils::osVersion,
        cores = unname(parallelly::availableCores()),
      modules = Sys.getenv("LOADEDMODULES")
  )
})
info <- value(f)
print(info)

future.batchtools documentation built on Oct. 19, 2026, 1:07 a.m.