pbdMPI (version 0.5-1)

global which, which.max, and which.min: Global Which Functions

Description

These functions are global which, which.max and which.min applying on distributed data for all ranks.

Usage

comm.which(x, arr.ind = FALSE, useNames = TRUE,
           comm = .pbd_env$SPMD.CT$comm)
comm.which.max(x, comm = .pbd_env$SPMD.CT$comm)
comm.which.min(x, comm = .pbd_env$SPMD.CT$comm)

Value

The global values (which(), which.max(), or which.min()) are returned to all ranks.

comm.which() returns with two columns, 'rank id' and 'index of

TRUE'.

comm.which.max() and comm.which.min() return with three values, 'the _smallest_ rank id', 'index of the _first_ maximum or minimum', and 'max/min value of x'.

Arguments

x

a 'logical' vector or array as in which(), or an 'numeric' objects in which.max() and which.min().

arr.ind

logical, as in which().

useNames

logical, as in which().

comm

a communicator number.

Author

Wei-Chen Chen , George Ostrouchov, Drew Schmidt, Pragneshkumar Patel, and Hao Yu.

Details

These functions will apply which(), which.max() and which.min() locally, and apply allgather() to get all local results from other ranks.

References

Programming with Big Data in R Website: https://pbdr.org/

See Also

comm.read.table()

Examples

if (FALSE) {
### Save code in a file "demo.r" and run with 2 processors by
### SHELL> mpiexec -np 2 Rscript demo.r

spmd.code <- "
### Initialize
suppressMessages(library(pbdMPI, quietly = TRUE))

if(comm.size() != 2){
  comm.cat(\"2 processors are requried.\n\", quiet = TRUE)
  finalize()
}

### Examples.
a <- 1:(comm.rank() + 1)

b <- comm.which(a == 2)
comm.print(b)
b <- comm.which.max(a)
comm.print(b)
b <- comm.which.min(a)
comm.print(b)

### Finish.
finalize()
"
# execmpi(spmd.code, nranks = 2L)
}