Tagged Questions

MPI is the Message Passing Interface, a library for distributed memory parallel programming and the de facto standard method for using distributed memory clusters for high-performance technical computing. Questions about using MPI for parallel programming go under this tag; questions on, eg, ...

learn more… | top users | synonyms

0
votes
0answers
2 views

Does PETSc really give speedup?

I searched linear solver library and found out PETSc library which considered to be powerful and useful library. PETSc consists implementations of various iterative methods with preconditioners and ...
0
votes
0answers
15 views

How to use MPI with QT?

Im trying to create an image processing program where I only have one window with buttons and what not. I need to use MPI on the heavy processing functions, like Fourier Transform and High/Low pass. ...
0
votes
1answer
35 views

MPI distributing array

I am trying to write a C++ program with MPI constructs. I want to read from a huge file and store the number into an array. I want the array to be local, i.e., I don't want all the threads to have the ...
1
vote
1answer
44 views

Shared memory access control mechanism for processes created by MPI

I have a shared memory used by multiple processes, these processes are created using MPI. Now I need a mechanism to control the access of this shared memory. I know that named semaphore and flock ...
0
votes
0answers
21 views

MPI doesn't work correctly with user interface

I'm trying to initialize Boost MPI in dll, using MS MPI from Microsoft Compute Cluster Pack. The dll itself works correctly with simple console project, but in a project with user interface on C#, ...
0
votes
1answer
12 views

Ok to call MPI_Isend multiple times on one buffer?

Regarding MPI_Isend, the MPI standard says "A nonblocking send call indicates that the system may start copying data out of the send buffer. The sender should not access any part of the send buffer ...
0
votes
0answers
14 views

MPICC: Two users on same cluster, if one user compiles it works, when the other user compiles it doesnt. What is wrong?

On our cluster we have recently ran into the following problem. We are running a helloworld type program which has been tested thoroughly. When the code is compiled with mpicc logged into account #1 ...
1
vote
2answers
36 views

Unbalanced load (v2.0) using MPI

(the problem is embarrassingly parallel) Consider an array of 12 cells: |__|__|__|__|__|__|__|__|__|__|__|__| and four (4) CPUs. Naively, I would run 4 parallel jobs and feeding 3 cells to each ...
0
votes
0answers
28 views

Use MPI with parallel HDF5

Hi I am beginner in HDF5 and MPI and I tried during several days before to run parallel hdf5 with MPI .So I run it with GFS2 and PVFS2 but I faced different problems that I can not overcome. So can ...
6
votes
0answers
67 views

Parallel application in python becomes much slower when using mpi rather than multiprocessing module

Lately I've observed a weird effect when I measured performance of my parallel application using the multiprocessing module and mpi4py as communication tools. The application performs evolutionary ...
0
votes
0answers
13 views

Better timer than mpi_Wtime

My mpi_wtime has resolution of 1e-6 seconds which is not good enough for me? What are pros and cons of using other timers like gettimeoftheday?
-1
votes
0answers
38 views

What algorithms can be implemented using MPI? [closed]

I read a textbook about MPI recently.But it does mention some practical usage of MPI, such as matrix multiplication and sieve of Eratosthenes. And i want to implement a algorithm using MPI. Are there ...
0
votes
1answer
18 views

Are OpenMPI and MPICH2 compatible?

I'm using OSX and bash. I am trying to run a program across two systems, and the host has OpenMPI 1.6.4 and the other node has MPICH2. I am running into problems, and this is my first time using any ...
0
votes
0answers
26 views

why is mpi_bcast so much slower than mpi_reduce?

Using MPI, we can do a broadcast to send an array to many nodes, or a reduce to combine arrays from many nodes onto one node. I guess that the fastest way to implement these will be using a binary ...
0
votes
1answer
24 views

Controlled premature termination of mpi program running under slurm?

I am running a script that does multiple subsequent mpirun calls through slurms squeue command. Each call to mpirun will write its output to an own directory, but there is a dependency between them in ...

1 2 3 4 5 … 95
15 30 50 per page