MPI_REDUCE_SCATTER(3)

📦

version 5.0.11 · 17 versions

🌈 MPI_REDUCE_SCATTER(3) — the Open MPI glow-up 💅

okay so bestie. you know how a reduce is like everyone shouting their answer into one big pot, and a scatter is like handing out slices of pizza? this is BOTH. at the same time. it's giving potluck-and-pizza-party. and yes, it comes in three iconic flavors:

MPI_Reduce_scatter — the classic, blocking, main character
MPI_Ireduce_scatter — the non-blocking one that's like "i'll text you back"
MPI_Reduce_scatter_init — the persistent one, sets things up with an info handle and never lets go

together they Combines values and scatters the results. periodt. 🎀

✨ SYNTAX ✨

💜 C Syntax

#include <mpi.h>

int MPI_Reduce_scatter(const void *sendbuf, void *recvbuf, const int recvcounts[],
     MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)

int MPI_Ireduce_scatter(const void *sendbuf, void *recvbuf, const int recvcounts[],
     MPI_Datatype datatype, MPI_Op op, MPI_Comm comm, MPI_Request *request)

int MPI_Reduce_scatter_init(const void *sendbuf, void *recvbuf,
     const int recvcounts[], 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_REDUCE_SCATTER(SENDBUF, RECVBUF, RECVCOUNTS, DATATYPE, OP,
             COMM, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER RECVCOUNTS(*), DATATYPE, OP, COMM, IERROR

MPI_IREDUCE_SCATTER(SENDBUF, RECVBUF, RECVCOUNTS, DATATYPE, OP,
             COMM, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER RECVCOUNTS(*), DATATYPE, OP, COMM, REQUEST, IERROR

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

🌸 Fortran 2008 Syntax

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

MPI_Ireduce_scatter(sendbuf, recvbuf, recvcounts, datatype, op, comm,
             request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: recvcounts(*)
     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_Reduce_scatter_init(sendbuf, recvbuf, recvcounts, datatype, op, comm,
             info, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: recvcounts(*)
     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 🍓

sendbuf — Starting address of send buffer (choice). it's where your stuff lives before the glow-up 💫
recvcounts — Integer array specifying the number of elements in result distributed to each process. Array must be identical on all calling processes. no cap, identical — like, everyone's copy has to match exactly or it's chaos 🔥
datatype — Datatype of elements of input buffer (handle).
op — Operation (handle). the reduction vibe, yk?
comm — Communicator (handle).
info — Info (handle, persistent).

🧚 OUTPUT PARAMETERS 🧚

recvbuf — Starting address of receive buffer (choice). your slice of the pizza lands here 🍕
request — Request (handle, non-blocking only). literally only for the non-blocking one, bestie, don't try it elsewhere 🙈
ierror — Fortran only: Error status (integer).

👑 DESCRIPTION 👑

So here's the tea, and the tea is two-step:

first, MPI_Reduce_scatter does an element-wise reduction on vector of count = S(i)recvcounts[i] elements in the send buffer defined by sendbuf, count, and datatype.

then, the resulting vector of results is split into n disjoint segments, where n is the number of processes in the group. Segment i contains recvcounts[i] elements. The ith segment is sent to process i and stored in the receive buffer defined by recvbuf, recvcounts[i], and datatype.

reduce. then scatter. it's literally in the name. icon behavior. ⭐

💅 USE OF IN-PLACE OPTION 💅

When the communicator is an intracommunicator, you can perform a reduce-scatter operation in-place (the output buffer is used as the input buffer). Use the variable MPI_IN_PLACE as the value of the sendbuf. In this case, the input data is taken from the top of the receive buffer. The area occupied by the input data may be either longer or shorter than the data filled by the output data.

yes bestie you can reuse the same buffer, and yes the sizes are allowed to be mismatched in either direction — "may be either longer or shorter" is right there in the text, don't come for me 💀

🌈 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🌈

When the communicator is an inter-communicator, the reduce-scatter operation occurs in two phases.

First, the result of the reduction performed on the data provided by the processes in the first group is scattered among the processes in the second group.
Then the reverse occurs: the reduction performed on the data provided by the processes in the second group is scattered among the processes in the first group.

For each group, all processes provide the same recvcounts argument, and the sum of the recvcounts values should be the same for both groups.

it's giving mutual exchange. both groups gotta bring the same receipts. 🫶

🔥 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 or silently skip the problem. 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.

The reason for this is the performance problems in ensuring that all collective routines return the same error value.

so like. don't expect a cute little error code back from these. it either nukes everything or pretends it saw nothing. heartbreaking but true 😭

💀 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. nuclear option fr fr 💀
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
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.

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.

bestie please internalize this one: MPI does not guarantee your program can keep going after an error. it might just. not. 😭

🎀 COPYRIGHT 🎀

2003-2026, The Open MPI Community

and that's the whole page, bestie!! 🦋✨ tysm for reading, this was the Open MPI MPI_REDUCE_SCATTER(3) page, dated Sep 11, 2026 — go forth and reduce-scatter responsibly, you absolute icon 👑💖

📦 where this page comes from

arch rolling: openmpi 5.0.11-1 · BSD-3-Clause AND LicenseRef-MPICH
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on 2026-11-06: a language model's rewrite of the original, under its terms