MPI_IALLTOALL(3)

📦
version 5.0.10 · 4 versions
, in ubuntu stonking

💖 MPI_IALLTOALL 💖

okay bestie, welcome to the Open MPI manual, and specifically to MPI_IALLTOALL(3) — the page where everybody talks to everybody and nobody gets left on read. it's giving group chat but make it parallel computing. let's go 💅

✨ NAME ✨

MPI_Alltoall, MPI_Ialltoall, MPI_Alltoall_init — All processes send data to all processes

like. literally all of them. to all of them. the whole friend group, no gatekeeping. 👑

✨ SYNTAX ✨

🎀 C Syntax 🎀

#include <mpi.h>

int MPI_Alltoall(const void *sendbuf, int sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, MPI_Comm comm)

int MPI_Ialltoall(const void *sendbuf, int sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, MPI_Comm comm, MPI_Request *request)

int MPI_Alltoall_init(const void *sendbuf, int sendcount,
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, MPI_Comm comm, MPI_Info info, MPI_Request *request)

🌸 Fortran Syntax 🌸

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_ALLTOALL(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
     RECVTYPE, COMM, IERROR)

     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE
     INTEGER COMM, IERROR

MPI_IALLTOALL(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
     RECVTYPE, COMM, REQUEST, IERROR)

     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE
     INTEGER COMM, REQUEST, IERROR

MPI_ALLTOALL_INIT(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNT,
     RECVTYPE, COMM, INFO, REQUEST, IERROR)

     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNT, SENDTYPE, RECVCOUNT, RECVTYPE
     INTEGER COMM, INFO, REQUEST, IERROR

🦋 Fortran 2008 Syntax 🦋

USE mpi_f08
MPI_Alltoall(sendbuf, sendcount, sendtype, recvbuf, recvcount, recvtype,
             comm, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
     TYPE(*), DIMENSION(..) :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Ialltoall(sendbuf, sendcount, sendtype, recvbuf, recvcount, recvtype,
             comm, request, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Alltoall_init(sendbuf, sendcount, sendtype, recvbuf, recvcount, recvtype,
                     comm, info, request, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: sendcount, recvcount
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Info), INTENT(IN) :: info
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

✨ INPUT PARAMETERS ✨

  • sendbuf — Starting address of send buffer (choice). where the gossip lives.
  • sendcount — Number of elements to send to each process (integer). "to each" is the whole vibe here, note that 📝
  • sendtype — Datatype of send buffer elements (handle).
  • recvcount — Number of elements to receive from each process (integer).
  • recvtype — Datatype of receive buffer elements (handle).
  • comm — Communicator over which data is to be exchanged (handle). the group chat server, basically.
  • info — Info (handle, persistent only)

✨ OUTPUT PARAMETERS ✨

  • recvbuf — Starting address of receive buffer (choice).
  • request — Request (handle, non-blocking only).
  • ierror — Fortran only: Error status (integer).

✨ DESCRIPTION ✨

okay so MPI_Alltoall is a collective operation where every process sends the same amount of data to every other process, and receives the same amount of data from every other process. mutual exchange. reciprocity. the healthy-relationship of MPI calls 💖

here's the whole thing as a picture — each process is doing 2n independent point-to-point communications, where n is the number of processes in communicator comm, and yes, that includes talking to itself, no that's not weird, it's self-care:

MPI_Comm_size(comm, &n);
for (i = 0, i < n; i++)
    MPI_Send(sendbuf + i * sendcount * extent(sendtype),
        sendcount, sendtype, i, ..., comm);
for (i = 0, i < n; i++)
    MPI_Recv(recvbuf + i * recvcount * extent(recvtype),
        recvcount, recvtype, i, ..., comm);

so each process chops its local sendbuf into n blocks — each block holding sendcount elements of type sendtype — and divides up its recvbuf the same way, except according to recvcount and recvtype. process j ships the k-th block of its local sendbuf to process k, and process k tucks that data into the j-th block of its local recvbuf. and ngl, the amount of data sent must equal the amount of data received, pairwise, between every pair of processes. no ghosting, no half-sends. 💀

🌈 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🌈

when the communicator is an inter-communicator, the gather operation happens in two phases. first, data is gathered from all the members of the first group and received by all the members of the second group. then (the plot twist) 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. it's giving mutual, it's giving both-ways, it's giving healthy communication ✨

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. no second act. one and done, slay. 💅

✨ USE OF IN-PLACE OPTION ✨

when the communicator is an intracommunicator, you can do an all-to-all operation in-place (the output buffer is used as the input buffer, iconic honestly — save that memory, queen). use the variable MPI_IN_PLACE as the value of sendbuf. in this case, sendcount 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 ✨

all arguments on all processes are significant. the comm argument especially — it must describe the same communicator on all processes. no mismatched servers in this group chat, bestie.

there are two MPI library functions that are more general than MPI_Alltoall:

  • MPI_Alltoallv lets you do all-to-all communication to and from buffers that need not be contiguous, and different processes may send and receive different amounts of data. flexible. unbothered.
  • MPI_Alltoallw takes MPI_Alltoallv's functionality and expands it to allow the exchange of data with different datatypes. the overachiever of the family 🌟

✨ ERRORS ✨

almost all MPI routines return an error value — C routines as the return result of the function, and Fortran routines in the last argument. noted. putting that in my journal. 📓

before the error value is returned, the current MPI error handler associated with the communication object (e.g., communicator, window, file) is called. if no communication object is associated with the MPI call, then the call is considered attached to MPI_COMM_SELF and will call the associated MPI error handler. when MPI_COMM_SELF is not initialized (i.e., before MPI_Init/MPI_Init_thread, after MPI_Finalize, or when using the Sessions Model exclusively) the error raises the initial error handler. the initial error handler can be changed by calling MPI_Comm_set_errhandler on MPI_COMM_SELF when using the World model, or the mpi_initial_errhandler CLI argument to mpiexec or info key to MPI_Comm_spawn/MPI_Comm_spawn_multiple. if no other appropriate error handler has been set, then the MPI_ERRORS_RETURN error handler is called for MPI I/O functions and the MPI_ERRORS_ABORT error handler is called for all other MPI functions.

Open MPI includes three predefined error handlers that can be used:

  • MPI_ERRORS_ARE_FATAL — Causes the program to abort all connected MPI processes. dramatic. nuclear. no survivors. 💀
  • MPI_ERRORS_ABORT — An error handler that can be invoked on a communicator, window, file, or session. When called on a communicator, it acts as if MPI_Abort was called on that communicator. If called on a window or file, acts as if MPI_Abort was called on a communicator containing the group of processes in the corresponding window or file. If called on a session, aborts only the local process. selective chaos, we love a targeted exit. 🎯
  • MPI_ERRORS_RETURN — Returns an error code to the application. the chill one. hands you the L and lets you deal with it. 🫶

MPI applications can also implement their own error handlers by calling:

  • MPI_Comm_create_errhandler then MPI_Comm_set_errhandler
  • MPI_File_create_errhandler then MPI_File_set_errhandler
  • MPI_Session_create_errhandler then MPI_Session_set_errhandler or at MPI_Session_init
  • MPI_Win_create_errhandler then MPI_Win_set_errhandler
note that MPI does not guarantee that an MPI program can continue past an error. fr fr. read that twice. 😭

see the MPI man page for a full list of MPI error codes.

see the Error Handling section of the MPI-3.1 standard for more information.

see also:

  • MPI_Alltoallv
  • MPI_Alltoallw

2003-2026, The Open MPI Community

and that's the whole page, bestie — you now know all-to-all better than most people know their own group chat dynamics. go forth and communicate symmetrically. you're the main character of this communicator 👑💖

with love from the Open MPI girlies, dated Sep 10, 2026 — mwah 💅✨🦋

📦 where this page comes from