SHMEM_N_PES(3)
okay so this is the Open MPI manual page SHMEM_N_PES(3), and itβs short but honestly iconic π
letβs go.
πΈ NAME πΈ
_num_pes, _num_pes, shmem_n_pes β returns the number of processing elements (PEs) used to run the application.
yes bestie, the NAME line really does list _num_pes twice β keeping every detail exactly as written, no editing, periodt π«Ά
β¨ SYNOPSIS β¨
C or C++:
#include <mpp/shmem.h>
int _num_pes (void)
int shmem_n_pes (void)
Fortran:
include 'mpp/shmem.fh'
I = NUM_PES ()
I = SHMEM_N_PES ()
π DESCRIPTION π
num_pes() or shmem_n_pes() return the total number of PEs running in an application.
Thatβs the whole tea: total PEs in the app, no cap β¨
π SEE ALSO
π COPYRIGHT π
2003-2026, The Open MPI Community
π¦ Signing off from the Open MPI manual/source, page SHMEM_N_PES(3), dated May 14, 2026 β bye bestie, go count those PEs! ππ
π¦ where this page comes from
- arch rolling: nvhpc-comm-libs 26.9-1 Β· LicenseRef-NVIDIA-HPC-SDK
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms