MPI_ALLTOALLV(3)

📦

version 5.0.10 · 7 versions

💅 okay so, `MPI_ALLTOALLV(3)` — Open MPI edition 💅

bestie. sit down. hydrate. we're doing the all-to-all where everyone has ✨opinions✨ about how much data they're sending.

this page covers three functions and they're all in the same glow-up family: MPI_Alltoallv, MPI_Ialltoallv, and MPI_Alltoallv_init. the official one-liner is:

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

which is a lot of words for "everybody talks to everybody, and nobody's talking the same amount." it's giving potluck where everyone brings a different number of cupcakes. 🧁

✨ SYNTAX ✨

🩷 C Syntax

#include <mpi.h>

int MPI_Alltoallv(const void *sendbuf, const int sendcounts[],
     const int sdispls[], 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[], MPI_Datatype sendtype,
     void *recvbuf, const int recvcounts[],
     const int rdispls[], MPI_Datatype recvtype, MPI_Comm comm,
     MPI_Request *request)

int MPI_Alltoallv_init(const void *sendbuf, const int sendcounts[],
     const int sdispls[], MPI_Datatype sendtype,
     void *recvbuf, const int recvcounts[],
     const int rdispls[], MPI_Datatype recvtype, MPI_Comm comm,
     MPI_Info info, MPI_Request *request)

🌸 Fortran Syntax

USE MPI
! or the older form: 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, COMM, REQUEST, IERROR)

     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNTS(*), SDISPLS(*), SENDTYPE
     INTEGER RECVCOUNTS(*), RDISPLS(*), RECVTYPE
     INTEGER COMM, REQUEST, IERROR

MPI_ALLTOALLV_INIT(SENDBUF, SENDCOUNTS, SDISPLS, SENDTYPE,
     RECVBUF, RECVCOUNTS, RDISPLS, RECVTYPE, COMM, INFO, REQUEST, IERROR)

     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNTS(*), SDISPLS(*), SENDTYPE
     INTEGER RECVCOUNTS(*), RDISPLS(*), RECVTYPE
     INTEGER COMM, INFO, REQUEST, IERROR

🦋 Fortran 2008 Syntax

USE mpi_f08
MPI_Alltoallv(sendbuf, sendcounts, sdispls, sendtype, recvbuf, recvcounts,
             rdispls, recvtype, comm, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
     TYPE(*), DIMENSION(..) :: recvbuf
     INTEGER, INTENT(IN) :: sendcounts(*), sdispls(*), recvcounts(*),
     rdispls(*)
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Ialltoallv(sendbuf, sendcounts, sdispls, sendtype, recvbuf, recvcounts,
             rdispls, recvtype, comm, request, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: sendcounts(*), sdispls(*),
     recvcounts(*), rdispls(*)
     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_Alltoallv_init(sendbuf, sendcounts, sdispls, sendtype, recvbuf, recvcounts,
                     rdispls, recvtype, comm, info, request, ierror)

     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: sendcounts(*), sdispls(*),
     recvcounts(*), rdispls(*)
     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 — where your send buffer starts. the home base of everything you're about to yeet. 📤
sendcounts — integer array, and entry i says how many elements you're sending to rank i. yes, you get a different number for literally every single rank. no cap.
sdispls — integer array, entry i = the displacement, aka the offset from sendbuf, measured in units of sendtype, where you start reading the stuff headed for rank i. 🎯
sendtype — the datatype of the elements in your send buffer.
recvcounts — integer array, entry j = how many elements you're receiving from rank j.
rdispls — integer array, entry j = the displacement, offset from recvbuf, measured in units of recvtype, where the data from rank j should land. 📥
recvtype — the datatype of the elements in your receive buffer.
comm — the communicator you're exchanging data over.
info — Info (handle, persistent only). 🫶

🎀 OUTPUT PARAMETERS 🎀

recvbuf — address of the receive buffer. this is where the goodies land.
request — Request (handle, non-blocking only). your lil ticket to track the async operation. 🎟️
ierror — Fortran only: error status.

🌈 DESCRIPTION 🌈

MPI_Alltoallv is a generalized collective operation where every process sends data to and receives data from all the other processes. it's MPI_Alltoall but with ✨flexibility✨ — you get to specify exactly what you send and receive, vector-style, via a displacement and an element count.

think of it like this: each process performs 2n independent point-to-point communications, where n is the number of processes in communicator comm, and yes that includes communication with itself. (bestie talks to herself, it's fine, it's valid.) the whole thing can be pictured as:

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);

so process j sends the k-th block of its local sendbuf to process k, and process k places that data into the j-th block of its local recvbuf. everyone's got their own little assigned neighborhood in there. 🏘️

when a pair of processes exchanges data, each one may pass different element count and datatype arguments — the only rule is that the sender specifies the same amount of data to send (in bytes) as the receiver expects to receive. the numbers don't have to match, the bytes do. periodt.

and 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. everybody's fed a different amount. iconic, honestly.

💌 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, 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 — meaning the output buffer is used as the input buffer. 🌟

to do it, use the variable MPI_IN_PLACE as the value of sendbuf. in that 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. seriously bestie. once. 💀

all arguments on all processes are significant. the comm argument, in particular, must describe the same communicator on all processes. everybody has to be on the same page or the whole thing falls apart.

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. different energy entirely, iykyk. 👑

💀 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.

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. brutal. effective. 😭
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.
MPI_ERRORS_RETURN — returns an error code to the application. the polite one. 🫶

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. ngl, it usually does not.

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 🌈

okay bestie, you've hit the end of the page and there's a whole little family of collectives out here living their best lives 🫶 — go give their manuals a read for the full lore:

MPI_Alltoallv
MPI_Ialltoallv
MPI_Alltoallv_init

these are literally the names this page ships with, so take them, keep them, no gatekeeping here 💅✨

and for the legal girlies in the back (iykyk), the whole paper trail is exactly one line, no more, no less:

okay that's a wrap, bestie 🎀 you now know every single thing this page had to say and honestly? iconic. this whole thing was written and blessed by the fine people over at The Open MPI Community, and it's dated Sep 10, 2026 — the exact stamp it wears is MPI_ALLTOALLV(3) 👑 so go forth and alltoallv responsibly. slay, and good luck out there 🌸🦋⭐

📦 where this page comes from

ubuntu stonking: openmpi-doc 5.0.10-6 · other
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-11-03: a language model's rewrite of the original, under its terms