Questions tagged [openmpi]

Open MPI is an open source implementation of the Message Passing Interface, a library for distributed memory parallel programming.

The Open MPI Project is an open-source implementation of the Message Passing Interface, a standardized and portable message-passing system designed to leverage to computational power of massively parallel, distributed memory computers.

Message passing is one of the distributed memory models most often used, with MPI being the most used message passing API, offering two types of communication between processes: point-to-point or collective. MPI can run in distributed and shared memory architectures.

An application using MPI consists usually of multiple simultaneously running processes, normally on different CPUs, which are able to communicate with each other. Normally, this type of application is programmed using the SPMD model. Nevertheless, most MPI implementations also support the MPMD model.

More information about the MPI standard may be found on the official MPI Forum website, in the official documentation, and in the Open MPI documentation.

1341 questions
0
votes
2 answers

Calling BLACS with more processes than used

I want to create a parallel program, which makes heavy use of SCALAPACK. The basis of SCALAPACK is BLACS, which itself relies on MPI for interprocess communication. I want to start the program with a defined number of processes (e.g. the number of…
Stefan
  • 2,460
  • 1
  • 17
  • 33
0
votes
1 answer

MPI: Receiving an already-received message?

I'm writing a C++ program that uses OpenMPI. It executes in "rounds", where in each round, process 0 sends chunks of data to the other processes, they do stuff to it and send results back, and when there are no more chunks to send, process 0 sends…
FrancesKR
  • 1,200
  • 1
  • 12
  • 27
0
votes
1 answer

MPI_Bcast hanging after some data transferred

I am running application that performs some transformations of large matrices on 7-node cluster. Nodes are connected via QDR 40 Gbit Infiniband. Open MPI 1.4.3 is installed on the system. Given matrix transformation requires large data exchange…
dzovejn
  • 3
  • 1
0
votes
1 answer

how to make amazon EC2 instances authenticate each other automatically?

I am using aws java sdk to launch EC2 instances (running Ubuntu 12.04) and run a distributed tool on them, the tool uses openMPI for message passing between the nodes and openMPI uses SSH to connect nodes with each other. The problem is that the…
Sami
  • 7,797
  • 18
  • 45
  • 69
0
votes
1 answer

infinite wait during openMPI run on a cluster of servers?

I have successfully set up the password less ssh between the servers and my computer. There is a simple openMPI program which is running well on the single computer. But ,unfortunately when i am trying this on a cluster ,neither i am getting a…
Ankur Gautam
  • 1,412
  • 5
  • 15
  • 27
0
votes
1 answer

Available cores vs. number of processes in openMPI

I tried the following "hello world" code, first on my system(8 cores) then on server(160 cores): int main(int argc, char *argv[]) { int numprocs, rank, namelen; char processor_name[MPI_MAX_PROCESSOR_NAME]; double t; t=MPI_Wtime(); …
Ankur Gautam
  • 1,412
  • 5
  • 15
  • 27
0
votes
0 answers

Open MPI, contact information is unknown

I am working on Mac OSX and using bash as my shell. I have been working for the past few hours trying to get the simplest of code to run using Open MPI on multiple computers. After fiddling with configuring Open MPI, I believe I am on the verge of…
Novice C
  • 1,344
  • 2
  • 15
  • 27
0
votes
2 answers

Unable to link ./oceanG to netcdff5.so

I am working on ROMS modeling. I just installed Openmpi on out linux cluster. I also adjusted LD_LIBRARY_PATH accordingly. I am now facing a unique probelm. when I give mpirun -np 4 ./oceanG ocean.in from bash shell, it runs the model fine.…
0
votes
1 answer

MPI/Pthread program does not scale

I have a MPI/Pthread program in which each MPI process will be running on a separate computing node. Within each MPI process, certain number of Pthreads (1-8) are launched. However, no matter how many Pthreads are launched within a MPI process, the…
xhe8
  • 419
  • 1
  • 5
  • 13
0
votes
1 answer

Linking C module to Fortran program while using OpenMPI: malloc causes segfault instead of null

I've got an old and messy Fortran program that uses MPI. It has one small module written in C, which tries to determine the largest allocatable block in the memory by calling malloc() iteratively until it returns null, and then returns a largest…
Serge Ts
  • 13
  • 3
0
votes
1 answer

How to install netcdf-openmpi-devel in Red Hat 6

I need to install netcdf-openmpi-devel on Red Hat 6. The problem is that this is not provided by the repositories I have: redhat and epel. I already tried downloading several fedora rpms, but for almost all of them, it's not possible to verify their…
John
  • 1
  • 2
0
votes
1 answer

Open MPI Hangs over multiple hosts

So I've been trying to get openmpi to work over two hosts, and whenever I type out my command it just hangs. Using the debug I managed to see that the native server successfully executed mpi after the /tmp command, however host2 hangs there. $…
ianhj0392
  • 1
  • 2
0
votes
1 answer

What effects does the mpicc/openmp flag -Dparallel have?

i ran into a problem with the flag -Dparallel (which we use in our code to enable parallel parts). When i use it, i dont get more then one OpenMP Thread anymore. Remark: We are running MPI/OpenMP hyrbrid code. A short code example: #include…
redimp
  • 1,061
  • 8
  • 12
0
votes
0 answers

mpi4py with OpenMPI: "ValueError: message: cannot guess count, datatype extent 0 (lb:0, ub:0)" in message_simple

I have mpi4py 1.3 installed with openMPI 1.6. Simple mpi4py demos like helloworld.py are working. In my code I call MPI.Comm.Allreduce(MPI.IN_PLACE, obj) , where obj is a numpy array. This results in the error message: ValueError: message: cannot…
Philipp Braun
  • 43
  • 1
  • 7
0
votes
1 answer

Set number of processors for openmpi

At the moment I am using something like this to run a c++ program: mpirun -np 4 ./test This works fine, but the problem is that I am wrapping the c++ program with Python. So I want to set the number of processors in Python and deliver it to the c++…
dolby
  • 643
  • 1
  • 8
  • 10