MPI_REDUCE_SCATTER_BLOCK(3)

📦

version 5.0.10 · 7 versions

💫 MPI_REDUCE_SCATTER_BLOCK(3) — the Open MPI glow-up

okay so bestie, welcome to the MPI_REDUCE_SCATTER_BLOCK(3) page from the Open MPI manual. this is one of those collective ops that sounds scary and is actually just "everyone mash their numbers together, then everybody gets a little slice" 🍰. we got three flavors here: MPI_Reduce_scatter_block, MPI_Ireduce_scatter_block, and MPI_Reduce_scatter_block_init. yep, non-blocking and persistent got invited to the party too 💅

✨ NAME ✨

MPI_Reduce_scatter_block, MPI_Ireduce_scatter_block, MPI_Reduce_scatter_block_init — combines values and scatters the results in blocks. that's the whole vibe: reduce it all, then hand out the blocks. 🎁

🌸 SYNTAX 🌸

C Syntax

#include <mpi.h>

int MPI_Reduce_scatter_block(const void *sendbuf, void *recvbuf, int recvcount,
     MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)

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

int MPI_Reduce_scatter_block_init(const void *sendbuf, void *recvbuf, int recvcount,
     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_BLOCK(SENDBUF, RECVBUF, RECVCOUNT, DATATYPE, OP,
             COMM, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER RECVCOUNT, DATATYPE, OP, COMM, IERROR

MPI_IREDUCE_SCATTER_BLOCK(SENDBUF, RECVBUF, RECVCOUNT, DATATYPE, OP,
             COMM, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER RECVCOUNT, DATATYPE, OP, COMM, REQUEST, IERROR

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

Fortran 2008 Syntax

USE mpi_f08
MPI_Ireduce_scatter_block(sendbuf, recvbuf, recvcount, datatype, op, comm,
             ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: recvcount
     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_block(sendbuf, recvbuf, recvcount, datatype, op, comm,
             request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: recvcount
     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_block_init(sendbuf, recvbuf, recvcount, datatype, op, comm,
             info, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN) :: recvcount
     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 the send buffer (choice). that's where your data starts, bestie.
recvcount — element count per block (non-negative integer). keep it non-negative, i don't make the rules 😤
datatype — datatype of elements of input buffer (handle).
op — operation (handle). the little function doing the mashing.
comm — communicator (handle).
info — info (handle, persistent only). so like, only MPI_Reduce_scatter_block_init gets to play with this one.

💖 OUTPUT PARAMETERS 💖

recvbuf — starting address of receive buffer (choice).
request — request (handle, non-blocking only).
ierror — Fortran only: error status (integer).

👑 DESCRIPTION 👑

okay so here's the tea ☕. MPI_Reduce_scatter_block first does an element-wise reduction on a vector of count = n * recvcount elements in the send buffer defined by sendbuf, count, and datatype, using the operation op, where n is the number of processes in the group of comm.

then — plot twist — that resulting vector of results gets split into n disjoint segments, where n is the number of processes in the group. each segment contains recvcount elements. the ith segment is sent to process i and stored in the receive buffer defined by recvbuf, recvcount, and datatype. so everybody gets exactly their lil block and nobody is left out. iconic behavior, honestly 🌈

🦋 USE OF IN-PLACE OPTION 🦋

when the communicator is an intracommunicator, you can do a reduce-scatter operation in-place — meaning the output buffer is used as the input buffer, no aux buffer required, so slay.

you do this by passing the variable MPI_IN_PLACE as the value of sendbuf. in that case, the input data is taken from the top of the receive buffer. and ngl, the area occupied by the input data may be either longer or shorter than the data filled by the output data. it's giving flexible 💅

🌟 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🌟

when the communicator is an inter-communicator, the reduce-scatter operation happens in two phases, bestie:

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. yes, both. don't be the one process that ruins it 🙈

💀 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. so 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. like, it'll just sit there. silently. 🫠

the reason for this is the performance problems in ensuring that all collective routines return the same error value. it's a whole thing.

😭 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. yeah. all of them. 💀
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. so like, don't count on it, fr fr.

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.

🎀 Copyright 🎀

2003-2026, The Open MPI Community

okay that's the whole page bestie, you're basically an MPI expert now, no cap 👑✨. this was dated Sep 10, 2026, and brought to you by 2003-2026, The Open MPI Community — huge thanks to them, they ate and left no crumbs 🍓💜 go forth and scatter those blocks, main character. periodt. 🦋

📦 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