Parallelize 'SimDesign' functions

Henrik Bengtsson

+ =

The futurize package allows you to easily turn sequential code into parallel code by piping the sequential code to the futurize() function. Easy!

TL;DR

library(futurize)
plan(multisession)
library(SimDesign)

res <- runSimulation(
  design = Design,
  replications = 1000,
  generate = Generate,
  analyse = Analyse,
  summarise = Summarise
) |> futurize()

Introduction

This vignette demonstrates how to use this approach to parallelize SimDesign functions such as runSimulation().

The SimDesign package provides a comprehensive framework for organizing Monte Carlo simulation experiments in R. It uses a structured generate-analyse-summarise workflow for designing, executing, and summarizing simulation studies. The replication-based nature of simulations makes them excellent candidates for parallelization.

Example: Monte Carlo simulation

The runSimulation() function runs Monte Carlo simulations over a design of experimental conditions. For example:

library(SimDesign)

Design <- createDesign(
  sample_size = c(10, 20, 40),
  distribution = c("norm", "chi")
)

Generate <- function(condition, fixed_objects) {
  N <- condition$sample_size
  dist <- condition$distribution
  if (dist == "norm") rnorm(N) else rchisq(N, df = 5)
}

Analyse <- function(condition, dat, fixed_objects) {
  c(mean_est = mean(dat))
}

Summarise <- function(condition, results, fixed_objects) {
  obs_bias <- bias(results[, "mean_est"],
    parameter = ifelse(condition$distribution == "norm", 0, 5))
  obs_RMSE <- RMSE(results[, "mean_est"],
    parameter = ifelse(condition$distribution == "norm", 0, 5))
  c(bias = obs_bias, RMSE = obs_RMSE)
}

res <- runSimulation(
  design = Design,
  replications = 100,
  generate = Generate,
  analyse = Analyse,
  summarise = Summarise
)

Here runSimulation() evaluates sequentially. To run in parallel, pipe to futurize():

library(futurize)
library(SimDesign)

res <- runSimulation(
  design = Design,
  replications = 100,
  generate = Generate,
  analyse = Analyse,
  summarise = Summarise
) |> futurize()

This will distribute the replications across the available parallel workers, given that we have set up parallel workers, e.g.

plan(multisession)

The built-in multisession backend parallelizes on your local computer and works on all operating systems. There are other parallel backends to choose from, including alternatives to parallelize locally as well as distributed across remote machines, e.g.

plan(future.mirai::mirai_multisession)

and

plan(future.batchtools::batchtools_slurm)

Supported Functions

The following SimDesign functions are supported by futurize():

• runSimulation()
• runArraySimulation()