openmpi

mpi4py freezes when calling Merge() and Disconnect()

吃可爱长大的小学妹 提交于 2021-02-11 09:58:47
问题 Why do Merge() and Disconnect() freeze when I try to use mpi4py on CentOS 7? I'm using Python 2.7.5, mpi4py 2.0.0, and I had to load the openmpi/gnu/1.8.8 module. I had trouble doing this under CentOS 6, and the only version of MPI that worked for me was openmpi/gnu/1.6.5 . Unfortunately, I don't see that version in the yum repositories for CentOS 7. Is there a way to trace what's happening in mpi4py or MPI? Is there a way to get the older version of MPI on CentOS 7? Here's the code I'm

mpi4py freezes when calling Merge() and Disconnect()

岁酱吖の 提交于 2021-02-11 09:53:22
问题 Why do Merge() and Disconnect() freeze when I try to use mpi4py on CentOS 7? I'm using Python 2.7.5, mpi4py 2.0.0, and I had to load the openmpi/gnu/1.8.8 module. I had trouble doing this under CentOS 6, and the only version of MPI that worked for me was openmpi/gnu/1.6.5 . Unfortunately, I don't see that version in the yum repositories for CentOS 7. Is there a way to trace what's happening in mpi4py or MPI? Is there a way to get the older version of MPI on CentOS 7? Here's the code I'm

brew install open-mpi fails

别来无恙 提交于 2021-02-10 23:14:54
问题 Trying to install openmpi , read that: Could not determine if REAL*16 bit-matches C type Ran brew doctor, and it's giving me: Your system is raring to brew full output % brew install open-mpi ==> Downloading http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6. Already downloaded: /Library/Caches/Homebrew/open-mpi-1.6.4.tar.bz2 ==> Using Homebrew-provided fortran compiler. This may be changed by setting the FC environment variable. ==> ./configure --prefix=/usr/local/Cellar/open

brew install open-mpi fails

半腔热情 提交于 2021-02-10 23:14:53
问题 Trying to install openmpi , read that: Could not determine if REAL*16 bit-matches C type Ran brew doctor, and it's giving me: Your system is raring to brew full output % brew install open-mpi ==> Downloading http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6. Already downloaded: /Library/Caches/Homebrew/open-mpi-1.6.4.tar.bz2 ==> Using Homebrew-provided fortran compiler. This may be changed by setting the FC environment variable. ==> ./configure --prefix=/usr/local/Cellar/open

brew install open-mpi fails

霸气de小男生 提交于 2021-02-10 23:13:02
问题 Trying to install openmpi , read that: Could not determine if REAL*16 bit-matches C type Ran brew doctor, and it's giving me: Your system is raring to brew full output % brew install open-mpi ==> Downloading http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6. Already downloaded: /Library/Caches/Homebrew/open-mpi-1.6.4.tar.bz2 ==> Using Homebrew-provided fortran compiler. This may be changed by setting the FC environment variable. ==> ./configure --prefix=/usr/local/Cellar/open

Intel Fortran to GNU Fortran Conversion [closed]

走远了吗. 提交于 2021-02-08 10:35:56
问题 Closed. This question needs debugging details. It is not currently accepting answers. Want to improve this question? Update the question so it's on-topic for Stack Overflow. Closed 4 years ago . Improve this question I am working on a custom CFD Solver written in Fortran 90 and MPI. The code contain 15+ Modules and was initially designed to work with the Intel Fortran compiler. Now since i do not have access to the Intel compiler I need to make it work using the GNU Fortran Compiler. I made

Intel Fortran to GNU Fortran Conversion [closed]

╄→гoц情女王★ 提交于 2021-02-08 10:33:26
问题 Closed. This question needs debugging details. It is not currently accepting answers. Want to improve this question? Update the question so it's on-topic for Stack Overflow. Closed 4 years ago . Improve this question I am working on a custom CFD Solver written in Fortran 90 and MPI. The code contain 15+ Modules and was initially designed to work with the Intel Fortran compiler. Now since i do not have access to the Intel compiler I need to make it work using the GNU Fortran Compiler. I made

NBody problem parallelization gives different results for the same input

牧云@^-^@ 提交于 2021-02-05 09:23:27
问题 This an MPI version of the NBody problem. I already have an OpenMP version and its results are the same as the nbody version with one thread, but the MPI results differ, above all at the last interactions. At the first interactions, the outputs are quite similar but at the end, the outputs differ a lot. #include <stdlib.h> #include <algorithm> #include <iostream> #include <fstream> #include <sstream> #include <cstring> #include <vector> #include <cstdlib> #include <chrono> #include <stdio.h>

Custom datatype (MPI_Datatype datatype)?

青春壹個敷衍的年華 提交于 2021-01-29 06:09:45
问题 Is there such a thing as a custom datatype in MPI, or do you have to flatten everything into a text string and pass as MPI_CHAR? If you are required to flatten everything, is there a built-in function I am overlooking? 回答1: The answer is MPI_Type_contiguous (the link is to the documentation). It allows you to block out a specific amount of space based on basic data types, and their respective offsets. 回答2: A much better answer is MPI_Type_create_struct. It allows you to replicate your struct

Correct way to source .bashrc for non-interactive shell

夙愿已清 提交于 2021-01-20 08:57:10
问题 I have been trying to resolve problems to be able to run openmpi on multiple nodes. Initially I had a problem with $PATH and $LD_LIBRARY_PATH variables not being updated from .bashrc file by openmpi session, so I manually added --prefix /path/to/openmpi to resolve this issue. Turns out that even the anaconda path variables are not being loaded as well. So ultimately I need ~/.bashrc file to be sourced from my home directory. How can I do that? Can anyone help me out please? UPDATE 01: I wrote