Open MPI logo

MPI_Alltoallv(3) man page (version 1.8.1)

  |   Home   |   Support   |   FAQ   |  

« Return to documentation listing

Table of Contents

Name

MPI_Alltoallv, MPI_Ialltoallv - All processes send different amount of data to, and receive different amount of data from, all processes

Syntax

C Syntax


#include <mpi.h>
int MPI_Alltoallv(const void *sendbuf, const int sendcounts[],
    const int sdispls]P, MPI_Datatype sendtype,
    void *recvbuf, const int recvcounts[],
    const int rdispls[], MPI_Datatype recvtype, MPI_Comm comm)
int MPI_Ialltoallv(const void *sendbuf, const int sendcounts[],
    const int sdispls]P, MPI_Datatype sendtype,
    void *recvbuf, const int recvcounts[],
    const int rdispls[], MPI_Datatype recvtype, MPI_Comm comm,
    MPI_Request *request)

Fortran Syntax


INCLUDE ’mpif.h’
MPI_ALLTOALLV(SENDBUF, SENDCOUNTS, SDISPLS, SENDTYPE,
    RECVBUF, RECVCOUNTS, RDISPLS, RECVTYPE, COMM, IERROR)
    <type>    SENDBUF(*), RECVBUF(*)
    INTEGER    SENDCOUNTS(*), SDISPLS(*), SENDTYPE
    INTEGER    RECVCOUNTS(*), RDISPLS(*), RECVTYPE
    INTEGER    COMM, IERROR
MPI_IALLTOALLV(SENDBUF, SENDCOUNTS, SDISPLS, SENDTYPE,
    RECVBUF, RECVCOUNTS, RDISPLS, RECVTYPE, REQUEST, COMM, IERROR)
    <type>    SENDBUF(*), RECVBUF(*)
    INTEGER    SENDCOUNTS(*), SDISPLS(*), SENDTYPE
    INTEGER    RECVCOUNTS(*), RDISPLS(*), RECVTYPE
    INTEGER    COMM, REQUEST, IERROR

C++ Syntax


#include <mpi.h>
void MPI::Comm::Alltoallv(const void* sendbuf,
    const int sendcounts[], const int displs[],
    const MPI::Datatype& sendtype, void* recvbuf,
    const int recvcounts[], const int rdispls[],
    const MPI::Datatype& recvtype)

Java Syntax


import mpi.*;
void MPI.COMM_WORLD.Alltoallv(Object sendbuf, int sendoffset, int sendcount[],
                              int sdispls[], MPI.Datatype sendtype,
                              Object recvbuf, int recvoffset, int recvcount[],
                              int rdispls[], MPI.Datatype recvtype)

Input Parameters

sendbuf
Starting address of send buffer.
sendoffset
Initial number of elements to skip at beginning of buffer (integer, Java-only).
sendcounts
Integer array, where entry i specifies the number of elements to send to rank i.
sdispls
Integer array, where entry i specifies the displacement (offset from sendbuf, in units of sendtype) from which to send data to rank i.
sendtype
Datatype of send buffer elements.
recvoffset
Initial number of elements to skip at beginning of buffer (integer, Java-only).
recvcounts
Integer array, where entry j specifies the number of elements to receive from rank j.
rdispls
Integer array, where entry j specifies the displacement (offset from recvbuf, in units of recvtype) to which data from rank j should be written.
recvtype
Datatype of receive buffer elements.
comm
Communicator over which data is to be exchanged.

Output Parameters

recvbuf
Address of receive buffer.
request
Request (handle, non-blocking only).
IERROR
Fortran only: Error status.

Description

MPI_Alltoallv is a generalized collective operation in which all processes send data to and receive data from all other processes. It adds flexibility to MPI_Alltoall by allowing the user to specify data to send and receive vector-style (via a displacement and element count). The operation of this routine can be thought of as follows, where each process performs 2n (n being the number of processes in communicator comm) independent point-to-point communications (including communication with itself).


    MPI_Comm_size(comm, &n);
    for (i = 0, i < n; i++)
        MPI_Send(sendbuf + sdispls[i] * extent(sendtype),
            sendcounts[i], sendtype, i, ..., comm);
    for (i = 0, i < n; i++)
        MPI_Recv(recvbuf + rdispls[i] * extent(recvtype),
            recvcounts[i], recvtype, i, ..., comm);

Process j sends the k-th block of its local sendbuf to process k, which places the data in the j-th block of its local recvbuf.

When a pair of processes exchanges data, each may pass different element count and datatype arguments so long as the sender specifies the same amount of data to send (in bytes) as the receiver expects to receive.

Note that process i may send a different amount of data to process j than it receives from process j. Also, a process may send entirely different amounts of data to different processes in the communicator.

WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR

When the communicator is an inter-communicator, the gather operation occurs in two phases. The data is gathered from all the members of the first group and received by all the members of the second group. Then the data is gathered from all the members of the second group and received by all the members of the first. The operation exhibits a symmetric, full-duplex behavior.

The first group defines the root process. The root process uses MPI_ROOT as the value of root. All other processes in the first group use MPI_PROC_NULL as the value of root. All processes in the second group use the rank of the root process in the first group as the value of root.

When the communicator is an intra-communicator, these groups are the same, and the operation occurs in a single phase.

Use of In-place Option

When the communicator is an intracommunicator, you can perform an all-to-all operation in-place (the output buffer is used as the input buffer). Use the variable MPI_IN_PLACE as the value of sendbuf. In this case, sendcounts, sdispls, and sendtype are ignored. The input data of each process is assumed to be in the area where that process would receive its own contribution to the receive buffer.

Notes

The specification of counts and displacements should not cause any location to be written more than once.

All arguments on all processes are significant. The comm argument, in particular, must describe the same communicator on all processes.

The offsets of sdispls and rdispls are measured in units of sendtype and recvtype, respectively. Compare this to MPI_Alltoallw, where these offsets are measured in bytes.

Errors

Almost all MPI routines return an error value; C routines as the value of the function and Fortran routines in the last argument. C++ functions do not return errors. If the default error handler is set to MPI::ERRORS_THROW_EXCEPTIONS, then on error the C++ exception mechanism will be used to throw an MPI::Exception object.

Before the error value is returned, the current MPI error handler is called. By default, this error handler aborts the MPI job, except for I/O function errors. The error handler may be changed with MPI_Comm_set_errhandler; the predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned. Note that MPI does not guarantee that an MPI program can continue past an error.

See Also


MPI_Alltoall
MPI_Alltoallw

« Return to documentation listing