Re: [OMPI users] slow MPI_BCast for messages size from 24K bytes to 800K bytes. (fwd)

2009-01-14 Thread Jeff Squyres
On Jan 13, 2009, at 3:32 PM, kmur...@lbl.gov wrote: With IB, there's also the issue of registered memory. Open MPI v1.2.x defaults to copy in/copy out semantics (with pre-registered memory) until the message reaches a certain size, and then it uses a pipelined register/RDMA protocol. Howe

[OMPI users] MPI flavor-agnostic libraries

2009-01-14 Thread Gregory D Abram
I'd like to release binary libraries that use MPI but are agnostic as to which flavor or MPI (eg. OpenMPI, MVAPICH, LAM...) that is used by the application linking the libraries. I've seen that there are some significant differences that stand in the way, for example that MPI_Comm is a pointer o

Re: [OMPI users] MPI flavor-agnostic libraries

2009-01-14 Thread Jeff Squyres
On Jan 14, 2009, at 9:49 AM, Gregory D Abram wrote: I'd like to release binary libraries that use MPI but are agnostic as to which flavor or MPI (eg. OpenMPI, MVAPICH, LAM...) that is used by the application linking the libraries. I've seen that there are some significant differences that s

Re: [OMPI users] MPI flavor-agnostic libraries

2009-01-14 Thread Mattijs Janssens
We did something similar in OpenFOAM by encapsulating our communications routines into a separate shared library which gets loaded through LD_LIBRARY_PATH. The api of this library does not expose any mpi specifics so for a new MPI version we only have to recompile our small communications libra

[OMPI users] mpirun (signal 15 Termination)

2009-01-14 Thread Hana Milani
Dear all, I have to run my code in parallel, therefore, I have installed openmpi-1.2.8 on a core2quad system with suse 11.0 linux and gfortran compiler. I have also downloaded blacs and scalapack from: http://www.netlib.org/scalapack/scalapack_installer.tgz. Everything has gone smoothly in ins

Re: [OMPI users] mpirun (signal 15 Termination)

2009-01-14 Thread Simon Hammond
Do you get anything else in the output? We sometimes get this but it means the execution has proceeded sucessfully. Si Hammond High Performance Systems Group Department of Computer Science University of Warwick 2009/1/14 Hana Milani : > Dear all, > > I have to run my code in parallel, therefore

Re: [OMPI users] mpirun (signal 15 Termination)

2009-01-14 Thread Hana Milani
Hello Simon, For running the program in parallel, I write: mpirun -np 4 ~/program output It takes a second that I receive the message: mpirun noticed that job rank 0 with PID 9477 on node linux-4pel exited on signal 15 (Terminated). and at the end of the output file, I receive: "3 addition

Re: [OMPI users] mpirun (signal 15 Termination)

2009-01-14 Thread Jeff Squyres
Signal 15 (i.e., SIGTERM) can also mean that a job scheduler killed you (e.g., if your job ran out of time). On Jan 14, 2009, at 11:46 AM, Simon Hammond wrote: Do you get anything else in the output? We sometimes get this but it means the execution has proceeded sucessfully. Si Hammond Hi

Re: [OMPI users] slow MPI_BCast for messages size from 24K bytes to 800K bytes. (fwd)

2009-01-14 Thread kmuriki
Hi Jeff, Here is the code with a warmup broadcast of 10K real values and actual broadcast of 100K real*8 values(different buffers): [kmuriki@n pub]$ more testbcast.f90 program em3d implicit real*8 (a-h,o-z) include 'mpif.h' ! em3d_inv main driver ! INITIALIZE MPI AND DETERMINE BOTH INDIVID

[OMPI users] MPI_COMM_WORLD not set in Fortran 90

2009-01-14 Thread David Robertson
Hello, I'm having a problem with MPI_COMM_WORLD in Fortran 90. I have tried with OpenMPI versions 1.2.6 and 1.2.8. Both versions are compiled with the PGI 8.0-2 suite. I've run the program in a debugger and MPI_COMM_WORLD is always 0, MPI_COMM_SELF is giving me a non-zero value though. The MP