hi,
I have a question on how LAM handles IO on a dual processors machine.
I wrote a small Fortran MPI program which reads/writes file on a NFS harddisk to exchange information between two processes. I have two same dual processors machines and I did two tests. first I tested the program on a single dual processors machine and ran two process on it. Then I tried the same program on two machines, and ran one process on each machine. the time that the program takes to run two process on a single machine is longer than that on two machines. is it normal? if not, what could be the reason? I suspect it has something to do with IO.
below is main part of the source code.
...
real, dimension(120000)::transferarray
call MPI_INIT( ierr )
call MPI_COMM_RANK( MPI_COMM_WORLD, myid, ierr )
call MPI_COMM_SIZE( MPI_COMM_WORLD, numprocs, ierr )
call MPI_GET_PROCESSOR_NAME(processor_name,namelen, ierr)
do n=1,2000
if (myid.ne.0) then
call MPI_RECV(temp,1,MPI_REAL,myid-1,9,MPI_COMM_WORLD,
& status,ierr)
open(unit=myid+200,file='transfer',status='old',
& recl=120000*4, form='UNFORMATTED',iostat=ierr)
rewind(myid+200)
read (myid+200) (transferarray(i),i=1,transfersize)
close(myid+200)
call MPI_BARRIER(MPI_COMM_WORLD,ierr)
end if
if (myid.eq.0) then
open(unit=myid+200,file='transfer',status='unknown',
& recl=120000*4, form='UNFORMATTED',iostat=ierr)
write (myid+200) (transferarray(i),i=1,transfersize)
call flush(myid+200)
close(myid+200)
call MPI_SEND(1.0,1,MPI_REAL,myid+1,9,
& MPI_COMM_WORLD,ierr)
call MPI_BARRIER(MPI_COMM_WORLD,ierr)
end if
end do
call MPI_BARRIER(MPI_COMM_WORLD,ierr)
call MPI_FINALIZE(ierr);
...
thanks
sg
|
