Parallelize 'Sim.DiffProc' 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(Sim.DiffProc)

# Define 1D SDE model
f <- expression(0)
g <- expression(1)
mod1d <- snssde1d(drift = f, diffusion = g, x0 = 1, M = 10, N = 100)
stat <- function(x, ...) mean(x)

res <- MCM.sde(mod1d, statistic = stat, R = 10, time = 0.5) |> futurize()

Introduction

This vignette demonstrates how to use this approach to parallelize Sim.DiffProc functions such as MCM.sde().

The Sim.DiffProc package provides a comprehensive framework for numerical simulation and inference of Stochastic Differential Equations (SDEs) in R. Because Monte Carlo simulation is highly iterative, running multiple replications in parallel can significantly reduce execution times.

Example: Monte Carlo simulation of SDEs

The MCM.sde() function performs Monte Carlo simulations for SDEs. For example:

library(Sim.DiffProc)

f <- expression(0)
g <- expression(1)
mod1d <- snssde1d(drift = f, diffusion = g, x0 = 1, M = 10, N = 100)
stat <- function(x, ...) mean(x)

res <- MCM.sde(mod1d, statistic = stat, R = 10, time = 0.5)

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

library(futurize)
library(Sim.DiffProc)

res <- MCM.sde(mod1d, statistic = stat, R = 10, time = 0.5) |> futurize()

This will distribute the Monte Carlo 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 Sim.DiffProc functions are supported by futurize():

• MCM.sde() with seed = TRUE as the default