This function creates a parallel processing cluster by detecting the number of available CPU cores on the machine, subtracting the number of cores specified to be left free, and using the remaining cores to create the cluster. This can be useful for speeding up computations that can be parallelized.
Value
A cluster object that can be used with parallel processing functions like parLapply and parApply.
Details
This function detects the total number of CPU cores available on the machine using the parallel package.
It then subtracts the number of cores specified in the free parameter and creates a cluster using the remaining cores.
On Linux systems, the function will use the snow package's getClusterOption() to determine the type of cluster
to create, ensuring compatibility with the system's parallel processing capabilities. On other operating systems,
the function defaults to creating a basic cluster without specifying a type, which should work for most cases.
Note that this function requires the parallel and snow packages to be installed and loaded.
Examples
if (FALSE) { # \dontrun{
# Create a cluster leaving 2 cores free for other tasks
cl <- createCluster(free = 2)
# Use the cluster for parallel processing
# ...
# Stop the cluster when done
parallel::stopCluster(cl)
} # }