mpi

问题 I don't understand how the environment is set on compute nodes when running with MPI under a scheduler. I do: mpirun -np 1 --hostfile ./hostfile foo.sh with foo.sh : #!/usr/bin/env zsh echo $LD_LIBRARY_PATH Then I do not recover the LD_LIBRARY_PATH I have got in an interactive shell... What are the initialization files that are executed/sourced at connection with MPI? note: I am under zsh, and I tried to put things in .zprofile or .zshenv instead of .zshrc, but it doesn't seem to make a

问题 I have a python script that recruits MPI for parallel calculations. The scheme of the calculations is following: data processing round 1 - data exchange between processes - data processing round 2. I have a 16 logical core machine (2 x Intel Xeon E5520 2.27GHz). For a reason round 1 cannot be run in parallel. Therefore, 15 cores stay idle. However, despite this fact calculations experience more than 2-fold slowdown. The problem is illustrated by this script (saved as test.py ): from mpi4py

问题 I have a python script that recruits MPI for parallel calculations. The scheme of the calculations is following: data processing round 1 - data exchange between processes - data processing round 2. I have a 16 logical core machine (2 x Intel Xeon E5520 2.27GHz). For a reason round 1 cannot be run in parallel. Therefore, 15 cores stay idle. However, despite this fact calculations experience more than 2-fold slowdown. The problem is illustrated by this script (saved as test.py ): from mpi4py

问题 I'm trying to run my unit tests through mpirun using ant. I have specified the task as: <target name="unitTest" depends="buildUnitTest"> <mkdir dir="reports"/> <junit fork="yes" jvm="mpirun java" printsummary="yes" haltonfailure="yes"> <classpath> <pathelement location="./bin"/> <pathelement location="/usr/share/java/junit4.jar"/> </classpath> <jvmarg value="-DDIM=3"/> <jvmarg value="-ea"/> <formatter type="plain"/> <batchtest todir="reports"> <fileset dir="test"> <include name="haparanda

问题 I'm trying to run my unit tests through mpirun using ant. I have specified the task as: <target name="unitTest" depends="buildUnitTest"> <mkdir dir="reports"/> <junit fork="yes" jvm="mpirun java" printsummary="yes" haltonfailure="yes"> <classpath> <pathelement location="./bin"/> <pathelement location="/usr/share/java/junit4.jar"/> </classpath> <jvmarg value="-DDIM=3"/> <jvmarg value="-ea"/> <formatter type="plain"/> <batchtest todir="reports"> <fileset dir="test"> <include name="haparanda

问题 The MPI standard, 3.0, says about mpi_scatterv: The specification of counts, types, and displacements should not cause any location on the root to be read more than once." However, my testing of mpi4py in python with the code below does not indicate that there is a problem with reading data from root more than once: import numpy as np from sharethewealth import sharethewealth comm = MPI.COMM_WORLD nprocs = comm.Get_size() rank = comm.Get_rank() counts = [16, 17, 16, 16, 16, 16, 15] displs =

问题 When my program works without MPI, then everything is fine with opening files, but when I add MPI, the files do not open. Why is that? My code: void fileEntry(string path, int n) { ofstream fout; fout.open(path); if (!fout.is_open()) { cout << "File open error"; } else { for(int i = 0; i < n; i++) { for(int j = 0; j < n; j++) { fout << rand() % 100 << " "; } fout << "\n"; } } fout.close(); } int main(int argc, char** argv) { MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &WORLD_RANK);

问题 Consider the following simple fortran program program test_vec_allocation use mpi implicit none integer(kind=8) :: N ! =========================BLACS and MPI======================= integer :: ierr, size, rank,dims(2) ! ------------------------------------------------------------- integer, parameter :: block_size = 100 integer :: context, nprow, npcol, local_nprow, local_npcol integer :: numroc, indxl2g, descmat(9),descvec(9) integer :: mloc_mat ,nloc_mat ,mloc_vec ,nloc_vec call blacs_pinfo

问题 consider the following piece of code using scalapack: ! if (norm2(h-x0) < tol) then tmp_vec = h - x0 call pdnrm2(N,norm,tmp_vec,1,1,descvec,1) if (norm < tol) then x=h converged = .true. exit endif s = r0 - alpha*v call pdgemv('N', N, N, 1.0, A, 1, 1, descmat, s, 1, 1, descvec, 1, 0.0,t, 1, 1, descvec, 1) its part of an iterative solver that i was trying, problem is that if my processor grid is two dimensional, my vectors do not have any elements on those procs, hence dnrm2 yields zero or the

问题 Can someone explain and tell me more about MPI_Comm_split communicator? MPI_Comm_split(MPI_COMM_WORLD, my_row, my_rank,&my_row_comm); This is just example i met by reading some basic documentations. Maybe someone could tell me how this communicator is working? 回答1: Just to begin with, let's have a look at the man page: MPI_Comm_split(3) MPI MPI_Comm_split(3) NAME MPI_Comm_split - Creates new communicators based on colors and keys SYNOPSIS int MPI_Comm_split(MPI_Comm comm, int color, int key,