MPIConfig.dat

2008-06-02T15:14:13Z

192.168.20.184:

The MPIConfig.dat file is no longer used in MPICH 1.2.x. You must write a command line with every option you intend to use.

----

The '''MPIConfig.dat''' is the setup file that distributes a [[MOHID Water]] simulation in several processors associating each nested domain to a processor.
Below an example explanation will be given, considering that:
# There is a simulation with a nested domain called '''Rias''' nested into a father domain called '''Galicia'''
# There are 2 computers connected through a network called PC1 and PC2, each with one processor available
# The simulation data files are all located in PC1
# The full path to the simulation father domain working folder is '''\\PC1\Applications\Galicia\exe'''
# PC1 has a virtual Mapped Network Drive with the letter '''T:\''' redirecting to the following full path '''\\PC1\Applications\'''.
# The whole simulation path is relative to this '''T:\''' drive.
# The MOHID executable compiled with MPI options is located in PC1
# The operative system being used is MS Windows XP Professional and the MPICH version installed is 1.1

The '''MPIConfig.dat''' must be configured like shown below. The construction of this file can be also done via [[MOHID GUI]].

'''Line 1''' - the first line contains a keyword '''exe''' containing the full path to the MOHID executable compiled with MPI options

exe \\PC1\EXE\Mohid_MPI.exe

'''Line 2''' - this line contains a keyword '''dir''' containing path to the father domain working folder ('''\exe''').

dir T:\Galicia\exe

'''Line 3''' - this line contains a keyword '''map''' containing the letter of the mapped drive '''T:''' followed by the full path to the father domain working folder ('''\exe''').

map T:\\PC1\Applications\Galicia\exe

'''Line 4''' - this line contains a keyword '''hosts'''. Below this line the definition of each computer specifying its name (e.g. PC1) and the number of processes to be launched in this computer.

hosts
PC1 1
PC2 1

For example, if PC1 is a computer carrying 2 processors the following definition could be written:

hosts
PC1 2

[[Category:MOHID Water]]

MPI

2007-12-12T17:44:59Z

192.168.20.184: /* Linux installation */

Is an acronym standing for '''Message Passing Interface'''. MPI are programing directives which allow a program to launch child processes in different ''nodes'' (other computers) through the network and communicate between them. This allows [[parallel processing]].
To Build a MOHID project for MPI refer to [[Compiling Mohid with MPI]].
==MPICH==
'''MPICH''' are the libraries containing the [[MPI]] directives.
===Linux installation===
*To build the '''MPICH''' libraries in linux, get them and:
>./configure FC=ifort --enable-f90
>make
>make testing
>make install
*To compile Mohid in linux in MPI refer to [[Compiling Mohid with MPI|this wiki]].
*To compile any program in linux simply type
>mpif90 -i-static foo.f90
*To use an MPI program on a Linux machine simply type these lines:
mpd &
mpiexec -n 4 ./MohidWater
mpdallexit
The three lines do the following: i) install the mpi daemon, ii) run MohidWater in 4 processes, iii)once the program is finished, kill all daemons.

==Samples==
===cpi===
This little C program calculates pi.
#include "mpi.h"
#include <stdio.h>
#include <math.h>

double f(double);

double f(double a)
{
return (4.0 / (1.0 + a*a));
}

int main(int argc,char *argv[])
{
int n, myid, numprocs, i;
double PI25DT = 3.141592653589793238462643;
double mypi, pi, h, sum, x;
double startwtime = 0.0, endwtime;
int namelen;
char processor_name[MPI_MAX_PROCESSOR_NAME];

MPI_Init(&argc,&argv);
MPI_Comm_size(MPI_COMM_WORLD,&numprocs);
MPI_Comm_rank(MPI_COMM_WORLD,&myid);
MPI_Get_processor_name(processor_name,&namelen);

fprintf(stdout,"Process %d of %d is on %s\n",
myid, numprocs, processor_name);
fflush(stdout);

n = 10000; /* default # of rectangles */
if (myid == 0)
startwtime = MPI_Wtime();
MPI_Bcast(&n, 1, MPI_INT, 0, MPI_COMM_WORLD);

h = 1.0 / (double) n;
sum = 0.0;
/* A slightly better approach starts from large i and works back */
for (i = myid + 1; i <= n; i += numprocs)
{
x = h * ((double)i - 0.5);
sum += f(x);
}
mypi = h * sum;

MPI_Reduce(&mypi, &pi, 1, MPI_DOUBLE, MPI_SUM, 0, MPI_COMM_WORLD);

if (myid == 0) {
endwtime = MPI_Wtime();
printf("pi is approximately %.16f, Error is %.16f\n",
pi, fabs(pi - PI25DT));
printf("wall clock time = %f\n", endwtime-startwtime);
fflush(stdout);
}

MPI_Finalize();
return 0;
}
=== f90pi===
The same little program in fortran 90:
!**********************************************************************
! pi3f90.f - compute pi by integrating f(x) = 4/(1 + x**2)
!
! (C) 2001 by Argonne National Laboratory.
! See COPYRIGHT in top-level directory.
!
! Each node:
! 1) receives the number of rectangles used in the approximation.
! 2) calculates the areas of it's rectangles.
! 3) Synchronizes for a global summation.
! Node 0 prints the result.
!
! Variables:
!
! pi the calculated result
! n number of points of integration.
! x midpoint of each rectangle's interval
! f function to integrate
! sum,pi area of rectangles
! tmp temporary scratch space for global summation
! i do loop index
!****************************************************************************
program main

use mpi

double precision PI25DT
parameter (PI25DT = 3.141592653589793238462643d0)

double precision mypi, pi, h, sum, x, f, a
integer n, myid, numprocs, i, rc
! function to integrate
f(a) = 4.d0 / (1.d0 + a*a)

call MPI_INIT( ierr )
call MPI_COMM_RANK( MPI_COMM_WORLD, myid, ierr )
call MPI_COMM_SIZE( MPI_COMM_WORLD, numprocs, ierr )
print *, 'Process ', myid, ' of ', numprocs, ' is alive'

sizetype = 1
sumtype = 2

do
if ( myid .eq. 0 ) then
write(6,98)
98 format('Enter the number of intervals: (0 quits)')
read(5,99) n
99 format(i10)
endif

call MPI_BCAST(n,1,MPI_INTEGER,0,MPI_COMM_WORLD,ierr)

! check for quit signal
if ( n .le. 0 ) exit

! calculate the interval size
h = 1.0d0/n

sum = 0.0d0
do i = myid+1, n, numprocs
x = h * (dble(i) - 0.5d0)
sum = sum + f(x)
enddo
mypi = h * sum

! collect all the partial sums
call MPI_REDUCE(mypi,pi,1,MPI_DOUBLE_PRECISION,MPI_SUM,0, &
MPI_COMM_WORLD,ierr)

! node 0 prints the answer.
if (myid .eq. 0) then
write(6, 97) pi, abs(pi - PI25DT)
97 format(' pi is approximately: ', F18.16, &
' Error is: ', F18.16)
endif

enddo
call MPI_FINALIZE(rc)
stop
end

==Other references==
*[[Parallel processing]].
*[[Setup a MOHID simulation using MPI]].
*[[Compiling Mohid with MPI]].
*[[MPIConfig.dat]].
[[Category:Technology]]
[[Category:Linux]]
[[Category:Windows]]

MohidWiki - User contributions [en]

MPIConfig.dat

MPI