RE: parallelization in Windows with gfortran
Not sure if your issue ever got fixed, I don't check this mailing-list very
often.
If using gfortran, you need to make sure you are updating the files in
\mpi\mpi_wing, not mpi_wini.
You should install a more recent version of MPICH and replace the files in the
mpi subfolders as Rik recommends. To check the health of MPI, in a cmd window
run:
smpd -install
mpiexec -register (filling out as required)
and then do something harmless like:
mpiexec -n 2 hostname
That should report the hostname twice if all is well.
Jeb Adams, Research Operations Manager
PKDM IS Liaison
~~~~~~~~~~~~~~~~~~
[email protected]
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Fisher Dennis
Sent: Monday, April 01, 2013 10:41 AM
To: [email protected]
Cc: Rik Schoemaker
Subject: [NMusers] parallelization in Windows with gfortran
Colleagues
I am configuring the -parafile option in Windows using gfortran. I am able to
run FPI but I am not successful with MPI. I installed 64-bit MPI using the
instructions suggested by Rik Schoemaker including copying fmpich2.lib to
mpi_wini. I am reasonably sure that my mpich installation is OK because when
I used a different passphrase during configuration of MPI, I got a message that
to the effect of "credentials failed". I then repeated the command mpiexec
-register with the correct passphrase.
Because I am using gfortran, I also copied fmpich2.lib to mpi_wing.
I also tried fmpich2g.lib in these locations.
Whenever I attempt to execute nonmem with:
nmfe72 foce_parallel.ctl outputfile -parafile=mpiwini8.pnm [nodes]=4
I get:
WARNINGS AND ERRORS (IF ANY) FOR PROBLEM 1
(WARNING 2) NM-TRAN INFERS THAT THE DATA ARE POPULATION.
CREATING MUMODEL ROUTINE...
1 file(s) copied.
Finished compiling fsubs
USING PARALLEL PROFILE mpiwini8.pnm
MPI TRANSFER TYPE SELECTED
Completed call to gfcompile.bat
Starting MPI version of nonmem execution ...
Could Not Find C:\nm72-gfortran\my.mpi\output
I tried various other pnm files (including the node3.pnm proposed by
Schoemaker, all yielding the same failure. Does anyone have any suggestions as
to how to make this work?
Dennis
Dennis Fisher MD
P < (The "P Less Than" Company)
Phone: 1-866-PLessThan (1-866-753-7784)
Fax: 1-866-PLessThan (1-866-753-7784)
www.PLessThan.com