Create a "parallel" cluster running sequentially in the current session

Description

The created cluster has only one node.

Usage

makeClusterSequential()

Details

Expression and function calls are evaluated in a local environment, inheriting the global environment.

Requirements

This function is only defined for R (>= 4.4.0).

Works in webR

Contrary to other cluster types, a makeClusterSequential() cluster works in webR, meaning you can use it for code that rely on the parallel package, e.g. y <- parLapply(cl, ...).

Examples

library(parallel)

cl <- makeClusterSequential()
print(cl)

y <- parLapply(cl, X = 1:3, fun = sqrt)
str(y)

pid <- Sys.getpid()
print(pid)
y <- clusterEvalQ(cl, Sys.getpid())
str(y)

abc <- 3.14
y <- clusterEvalQ(cl, { abc <- 42; abc })
str(y)
stopifnot(abc == 3.14)

clusterExport(cl, "abc", envir = environment())
rm(abc)
y <- clusterEvalQ(cl, { abc })
str(y)
stopifnot(y[[1]] == 3.14)