Set the number of threads to use for parallel processing in Seurat

Source: R/utilities.R

Set the number of threads to use for parallel processing in Seurat

setThreads(n = NULL, verbose = TRUE)

Arguments

n

Number of threads (>= 1) to use for processing. If NULL, will be set to (available cores / 2).

verbose

Whether to print the number of threads set and available cores to the console

Details

The default is one thread. Calling setThreads() without specifying n selects half the available cores. This setting applies to supported C++ kernels in NormalizeData, FindVariableFeatures, ScaleData, SCTransform, RunPCA, FindNeighbors, FindClusters, and CCA/RPCA integration. RunUMAP also passes this setting to the uwot backend. Support depends on the input type, backend, and arguments; not every stage of these functions can be run in parallel.

Note that this thread setting is separate from plan.

See also

getThreads