openmpi | an open source java 3d secure mpi implementation

while compiling I get the following warning:

/usr/bin/ld: warning: libgfortran.so.3, needed by /usr/openmpi-4.0.3rc4/lib64/libmpi_usempi.so, may conflict with libgfortran.so.5

It does create the .exe but when executing it an error occurs:

ideal.exe: error while loading shared libraries: libgfortran.so.5: cannot open shared object file: No such file or directory

I search for it to try and link it but it didn't work

whereis libgfortran.so.5

libgfortran.so: /usr/lib64/libgfortran.so.3

I don't have much knowlegde about linux or compilers and I'm working on a SUSE server without sudo permission. The gnu fortran compiler I'm using is in my home directory /home/gomezmr/gcc . Does anyone know how to solve this? Thank you.

I have an array of type Matrix structs which the program got from user's input. I need to distribute the matrices to processes with OpenMPI. I tried using Scatter but I am quite confused about the arguments needed for the program to work (and also how to receive the data in each local arrays). Here is my current code:

I'm facing a strange problem when trying to use MPI on a cluster, which uses Slurm as a job scheduler. On both cases, I'm trying to run this simple python program:

I am trying to start a sequence of MPI_Igather calls (non-blocking collectives from MPI 4), do some work, and then whenever an Igather finishes, do some more work on that data.

That works fine, unless I start the Igather's from different threads on each MPI rank. In that case I often get a deadlock, even though I call MPI_Init_thread to make sure that MPI_THREAD_MULTIPLE is provided. Non-blocking collectives do not have a tag to match sends and receives, but I thought this is handled by the MPI_Request object associated with each collective operation?

The most simple example I found failing is this:

• given np MPI ranks, each of which has a local array of length np
• start one MPI_Igather for each element i, gathering those elements on process i.
• the i-loop is parallelized using OpenMP
• then call MPI_Waitall() to finish all communication. This is where the program hangs when setting OMP_NUM_THREADS to a value larger than 1.

I made two variants of this program: igather_threaded.cpp (the code below), which behaves as described above, and igather_threaded_v2.cpp, which gathers everything on MPI rank 0. This version does not deadlock, but the data is not ordered correctly either.

igather_threaded.cpp:

I have a program that I would like to compile with CMake + make but using two different MPI distributions, OpenMPI and MPICH.

In Ubuntu, I have both installed; these are all the compiler wrappers I have installed:

The following code fails during runtime because of the MPI Scatter Error which I am not able to fix. When following documentation and other similar error pages, I didn't see any issue. Please help. I am using openmpi/4.0.5-gcc.

I have raster bricks for daily minimum and maximum temperatures from Daymet. The geographic extent is continental U.S. Each brick has 365 layers, one for each day of the year. I want to calculate the mean daily temperature from pair of min. and max. bricks by year. It seems the overlay() function in the R raster package does what I want. For example:

I have a code using Fortran modules. I can build it with no problems under normal circumstances. CMake takes care of the ordering of the module files.

However, using a gitlab runner, it SOMETIMES happens that cmake does NOT order the Fortran modules by dependencies, but alphabetically instead, which than leads to a build failure.

The problem seems to occur at random. I have a branch that built in the CI. After adding a commit, that modified a utility script not involved in any way in the build, I ran into this problem. There is no difference in the output of the cmake configure step.

I use the matrix configuration for the CI to test different configurations. I found, that I could trigger this by adding another mpi version (e.g. openmpi/4.1.6). Without that version, it built. With it added in the matrix, ALL configurations showed the problem.

I am using Conda on a shared compute cluster where numerical and io libraries have been tune for the system. How can I tell Conda to use these and only worry about the libraries and packages which are not already there on path?

For example:

There is a openmpi library installed and the package which I would like to install and mange with Conda has it also as a dependency. How can I tell Conda to just worry about what is not there?