xx_comm_range: Global Range, Max, and Min Functions

In pbdMPI: R Interface to MPI for HPC Clusters (Programming with Big Data Project)

Global Range, Max, and Min Functions

Description

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

Usage

comm.range(..., na.rm = FALSE, comm = .pbd_env$SPMD.CT$comm)
comm.max(..., na.rm = FALSE, comm = .pbd_env$SPMD.CT$comm)
comm.min(..., na.rm = FALSE, comm = .pbd_env$SPMD.CT$comm)

Arguments

...	an 'numeric' objects.
na.rm	if NA removed or not.
comm	a communicator number.

Details

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

Value

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

Author(s)

Wei-Chen Chen [e-mail], George Ostrouchov, Drew Schmidt, Pragneshkumar Patel, and Hao Yu.

References

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

Examples

## Not run: 
### 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.range(a)
comm.print(b)
b <- comm.max(a)
comm.print(b)
b <- comm.min(a)
comm.print(b)

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

## End(Not run)

pbdMPI documentation built on Sept. 18, 2024, 9:07 a.m.