ae_alltoall: All to All

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

All to All

Description

These functions make calls to MPI_Alltoall() and MPI_Alltoallv().

Usage

spmd.alltoall.integer(x.send, x.recv, send.count, recv.count,
                      comm = .pbd_env$SPMD.CT$comm)
spmd.alltoall.double(x.send, x.recv, send.count, recv.count,
                    comm = .pbd_env$SPMD.CT$comm)
spmd.alltoall.raw(x.send, x.recv, send.count, recv.count,
                  comm = .pbd_env$SPMD.CT$comm)

spmd.alltoallv.integer(x.send, x.recv, send.count, recv.count,
                       sdispls, rdispls, comm = .pbd_env$SPMD.CT$comm)
spmd.alltoallv.double(x.send, x.recv, send.count, recv.count,
                      sdispls, rdispls, comm = .pbd_env$SPMD.CT$comm)
spmd.alltoallv.raw(x.send, x.recv, send.count, recv.count,
                   sdispls, rdispls, comm = .pbd_env$SPMD.CT$comm)

Arguments

x.send	an object to send.
x.recv	an object to receive
send.count	send counter
recv.count	recv counter
sdispls	send dis pls
rdispls	recv dis pls
comm	a communicator number.

Details

These are very low level functions. Use with cautions. Neigher S4 method nor long vector is supported.

Value

These are very low level functions. Use with cautions. Neigher S4 method nor long vector is supported.

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/

See Also

allgather(), allgatherv().

Examples

## Not run: 
### Save code in a file "demo.r" and run with 2 processors by
### SHELL> mpiexec -np 2 Rscript --vanilla [...].r

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

### Examples.
n <- as.integer(2)
x <- 1:(.comm.size * n)
comm.cat(\"Original x:\n\", quiet = TRUE)
comm.print(x, all.rank = TRUE)

x <- as.integer(x)
y <- spmd.alltoall.integer(x, integer(length(x)), n, n)
comm.cat(\"\nAlltoall y:\n\", quiet = TRUE)
comm.print(y, all.rank = TRUE)

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

## End(Not run)

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