Skip to contents

Parallelize 'BiocParallel' functions

Henrik Bengtsson

Source: vignettes/futurize-61-BiocParallel.md
futurize-61-BiocParallel.Rmd

The Bioconductor 'BiocParallel' image+ The 'futurize' hexlogo= The 'future' logo

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

You can use futurize to make BiocParallel functions to parallelize via any of the [parallel backends] supported by Futureverse, e.g.

library(futurize)
plan(multisession)
library(BiocParallel)

slow_fcn <- function(x) {
  Sys.sleep(0.1)  # emulate work
  x^2
}

xs <- 1:1000
ys <- bplapply(xs, slow_fcn) |> futurize()

Introduction

This vignette demonstrates how to use this approach to parallelize functions such as bplapply(), bpmapply(), and bpvec() in the BiocParallel package. For example, consider the bplapply() function. It works like base-R lapply(), but uses the BiocParallel framework to process the tasks concurrently. It is commonly used something like:

library(BiocParallel)
xs <- 1:1000
ys <- bplapply(xs, slow_fcn)

The parallel backend is controlled by the BiocParallel::register(), similar to how we use future::plan() in futureverse. We can use the futurize package to tell BiocParallel to hand over the orchestration of parallel tasks to futureverse. All we need to do is to pass the expression to futurize() as in:

library(futurize)
library(BiocParallel)
xs <- 1:1000
ys <- bplapply(xs, slow_fcn) |> futurize()

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

plan(multisession)

The built-in multisession backend parallelizes on your local computer and it 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 futurize() function supports parallelization of all BiocParallel functions that take argument BPPARAM. Specifically,

The following functions are currently not supported: