MPI_ALLREDUCE_INIT(3)

📦

version 5.0.10 · 4 versions

✨ NAME ✨

okay so this page is about a whole trio of besties: MPI_Allreduce, MPI_Iallreduce, and MPI_Allreduce_init — and what they do is combine values from all processes and distribute the result back to all processes. every single one of them. no cap. 💖

🌸 SYNTAX 🌸

C Syntax

#include <mpi.h>

int MPI_Allreduce(const void *sendbuf, void *recvbuf, int count,
                  MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)

int MPI_Iallreduce(const void *sendbuf, void *recvbuf, int count,
                   MPI_Datatype datatype, MPI_Op op, MPI_Comm comm,
                   MPI_Request *request)

int MPI_Allreduce_init(const void *sendbuf, void *recvbuf, int count,
                       MPI_Datatype datatype, MPI_Op op, MPI_Comm comm,
                       MPI_Info info, MPI_Request *request)

Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_ALLREDUCE(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER COUNT, DATATYPE, OP, COMM, IERROR

MPI_IALLREDUCE(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER COUNT, DATATYPE, OP, COMM, REQUEST, IERROR

MPI_ALLREDUCE_INIT(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, INFO, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER COUNT, DATATYPE, OP, COMM, INFO, REQUEST, IERROR

Fortran 2008 Syntax

USE mpi_f08
MPI_Allreduce(sendbuf, recvbuf, count, datatype, op, comm, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
     TYPE(*), DIMENSION(..) :: recvbuf
     INTEGER, INTENT(IN) :: count
     TYPE(MPI_Datatype), INTENT(IN) :: datatype
     TYPE(MPI_Op), INTENT(IN) :: op
     TYPE(MPI_Comm), INTENT(IN) :: comm
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Iallreduce(sendbuf, recvbuf, count, datatype, op, comm, request,
             ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: count
     TYPE(MPI_Datatype), INTENT(IN) :: datatype
     TYPE(MPI_Op), INTENT(IN) :: op
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Allreduce_init(sendbuf, recvbuf, count, datatype, op, comm, info, request,
             ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: count
     TYPE(MPI_Datatype), INTENT(IN) :: datatype
     TYPE(MPI_Op), INTENT(IN) :: op
     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 🫶

the stuff you gotta hand over, bestie:

sendbuf — where your send buffer starts (it's a choice, so go wild). 💌
count — how many elements are in the send buffer (integer, obviously).
datatype — the datatype of the elements in the send buffer (handle).
op — the operation you wanna run (handle).
comm — the communicator (handle).
info — info (handle, persistent only). like, this one's only for the persistent girlies.

💅 OUTPUT PARAMETERS 💅

the stuff it hands back to you:

recvbuf — where your receive buffer starts (also a choice).
request — the request (handle, non-blocking only). only the async girlies get one of these.
ierror — Fortran only: error status (integer).

👑 DESCRIPTION 👑

It's literally the same as MPI_Reduce (that's the #mpi-reduce link, iykyk) except that the result appears in the receive buffer of all the group members. not just the one, not just the root — everybody eats. 🍓

Example 1: A routine that computes the product of a vector and an array that are distributed across a group of processes and returns the answer at all nodes (compare with Example 2, with MPI_Reduce, below).

SUBROUTINE PAR_BLAS2(m, n, a, b, c, comm)
REAL a(m), b(m,n)    ! local slice of array
REAL c(n)            ! result
REAL sum(n)
INTEGER n, comm, i, j, ierr

! local sum
DO j= 1, n
  sum(j) = 0.0
  DO i = 1, m
    sum(j) = sum(j) + a(i)*b(i,j)
  END DO
END DO

! global sum
CALL MPI_ALLREDUCE(sum, c, n, MPI_REAL, MPI_SUM, comm, ierr)

! return result at all nodes
RETURN

Example 2: A routine that computes the product of a vector and an array that are distributed across a group of processes and returns the answer at node zero.

SUBROUTINE PAR_BLAS2(m, n, a, b, c, comm)
REAL a(m), b(m,n)    ! local slice of array
REAL c(n)            ! result
REAL sum(n)
INTEGER n, comm, i, j, ierr

! local sum
DO j= 1, n
  sum(j) = 0.0
  DO i = 1, m
    sum(j) = sum(j) + a(i)*b(i,j)
  END DO
END DO

! global sum
CALL MPI_REDUCE(sum, c, n, MPI_REAL, MPI_SUM, 0, comm, ierr)

! return result at node zero (and garbage at the other nodes)
RETURN

🦋 USE OF IN-PLACE OPTION 🦋

When the communicator is an intracommunicator, you can perform an all-reduce operation in-place — that means the output buffer is used as the input buffer. To do that, use the variable MPI_IN_PLACE as the value of sendbuf at all processes. 💫

Not me about to use MPI_IN_PLACE without reading the fine print — note that MPI_IN_PLACE is a special kind of value; it has the same restrictions on its use as MPI_BOTTOM.

Because the in-place option converts the receive buffer into a send-and-receive buffer, a Fortran binding that includes INTENT must mark these as INOUT, not OUT. okay? periodt. 💅

🌈 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🌈

When the communicator is an inter-communicator, the reduce operation occurs in two phases. The data is reduced from all the members of the first group and received by all the members of the second group. Then the data is reduced 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, balanced, iconic. ✨

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

💀 NOTES ON COLLECTIVE OPERATIONS 💀

The reduction functions (MPI_Op) do not return an error value. As a result, if the functions detect an error, all they can do is either call MPI_Abort (the #mpi-abort link) or silently skip the problem. so like… yeah. 🙈

Thus, if you change the error handler from MPI_ERRORS_ARE_FATAL to something else, for example MPI_ERRORS_RETURN, then no error may be indicated. screaming, crying, throwing up — but quietly.

😭 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 — that's the #mpi-init / #mpi-init-thread links — 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 (link: #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 (links: #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. 💀
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.

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

MPI_Comm_create_errhandler then MPI_Comm_set_errhandler (links: #mpi-comm-create-errhandler, #mpi-comm-set-errhandler)
MPI_File_create_errhandler then MPI_File_set_errhandler (links: #mpi-file-create-errhandler, #mpi-file-set-errhandler)
MPI_Session_create_errhandler then MPI_Session_set_errhandler or at MPI_Session_init (links: #mpi-session-create-errhandler, #mpi-session-set-errhandler, #mpi-session-init)
MPI_Win_create_errhandler then MPI_Win_set_errhandler (links: #mpi-win-create-errhandler, #mpi-win-set-errhandler)

Note that MPI does not guarantee that an MPI program can continue past an error. 🫠

See the MPI man page (link: #open-mpi) for a full list of MPI error codes (link: #open-mpi-errors).

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

🎀 Copyright 🎀

2003-2026, The Open MPI Community

and that's the whole vibe, bestie 💖 this page is from Open MPI, dated Sep 10, 2026 — go forth and all-reduce, you absolute main character 👑🦋

📦 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