I am a first time user of NETCDF, so this may be in novice question.
The question is specifically about the example simple_xy_par_wr.f90, in
particular about the following section:
--------------------
! The dimids array is used to pass the IDs of the dimensions of
! the variables. Note that in fortran arrays are stored in
! column-major format.
dimids = (/ y_dimid, x_dimid /)
--------------------
The question is about whether or not the line above is correct, or the X
and Y have to be flipped? The reason for this question is the following.
Instead of varying the global grid size depending on the number of
processors as is done in the example above, I wanted to create a simple
example that had a fixed global array size that could be run using
different MPI ranks.
The only way I could get it to work was if I switched the dimensions
corresponding to the line above. I have included this new example. Since
very few lines have changed, you can actually do a diff between this code
and the original example.
Can you please let me know if I'm doing something wrong in my code, or is
the example above is incorrect and works only because the array in question
was a 1D array?
Thanks!
--rr
Please Note: Email transcribed by speech recognition; May have unintended
words in some cases, especially when I fail to do a good job of
proofreading!
! This is part of the netCDF package.
! Copyright 2006 University Corporation for Atmospheric
Research/Unidata.
! See COPYRIGHT file for conditions of use.
! This is a very simple example which writes a 2D array of sample
! data. To handle this in netCDF we create two shared dimensions,
! "x" and "y", and a netCDF variable, called "data". It uses
! parallel I/O to write the file from all processors at the same
! time.
! This example demonstrates the netCDF Fortran 90 API. This is part
! of the netCDF tutorial, which can be found at:
! http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-tutorial
! Full documentation of the netCDF Fortran 90 API can be found at:
! http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-f90
! $Id: simple_xy_par_wr.f90,v 1.3 2010/06/01 15:34:49 ed Exp $
program simple_xy_par_wr
use netcdf
implicit none
include 'mpif.h'
! This is the name of the data file we will create.
character (len = *), parameter :: FILE_NAME = "simple_xy_par-rr.nc"
! We are writing 2D data.
integer, parameter :: NDIMS = 2
! When we create netCDF files, variables and dimensions, we get back
! an ID for each one.
integer :: ncid, varid, dimids(NDIMS)
integer :: x_dimid, y_dimid
! These will tell where in the data file this processor should
! write.
integer :: start(NDIMS), count(NDIMS)
! This is the data array we will write. It will just be filled with
! the rank of this processor.
integer, allocatable :: data_out(:,:)
! MPI stuff: number of processors, rank of this processor, and error
! code.
integer :: p, my_rank, ierr
! Loop indexes, and error handling.
integer :: x, y, stat
integer, parameter :: NR=5, NC=64 ! global array size
integer :: NCL ! local # cols
! Initialize MPI, learn local rank and total number of processors.
call MPI_Init(ierr)
call MPI_Comm_rank(MPI_COMM_WORLD, my_rank, ierr)
call MPI_Comm_size(MPI_COMM_WORLD, p, ierr)
NCL = NC / p
! Create some pretend data. We just need one row.
allocate(data_out(NR,NCL), stat = stat)
if (stat .ne. 0) stop 3
do y = 1, NCL
do x = 1, NR
data_out(x,y) = my_rank*100000 + (y-1)*NR + x
end do
end do
! Create the netCDF file. The NF90_NETCDF4 flag causes a
! HDF5/netCDF-4 file to be created. The comm and info parameters
! cause parallel I/O to be enabled. Use either NF90_MPIIO or
! NF90_MPIPOSIX to select between MPI/IO and MPI/POSIX.
call check(nf90_create(FILE_NAME, IOR(NF90_NETCDF4, NF90_MPIIO),
ncid, &
comm = MPI_COMM_WORLD, info = MPI_INFO_NULL))
! Define the dimensions. NetCDF will hand back an ID for
! each. Metadata operations must take place on all processors.
call check(nf90_def_dim(ncid, "x", NR, x_dimid))
call check(nf90_def_dim(ncid, "y", NC, y_dimid)) ! global ?????
!--------- WARNING: The comment below is from original code and WRONG
------!
! ! The dimids array is used to pass the IDs of the dimensions of
! ! the variables. Note that in fortran arrays are stored in
! ! column-major format.
! dimids = (/ y_dimid, x_dimid /) !!! Example was wrong-rreddy???
!------------------------------------------------------------------------
dimids = (/ x_dimid, y_dimid /)
! Define the variable. The type of the variable in this case is
! NF90_INT (4-byte integer).
call check(nf90_def_var(ncid, "data", NF90_INT, dimids, varid))
! End define mode. This tells netCDF we are done defining
! metadata. This operation is collective and all processors will
! write their metadata to disk.
call check(nf90_enddef(ncid))
! Write the pretend data to the file. Each processor writes one row.
start = (/ 1, my_rank*NCL + 1/)
count = (/ NR, NCL /)
call check(nf90_put_var(ncid, varid, data_out, start = start, &
count = count))
! Close the file. This frees up any internal netCDF resources
! associated with the file, and flushes any buffers.
call check( nf90_close(ncid) )
! Free my local memory.
deallocate(data_out)
if (my_rank .eq. 0) print *, "*** SUCCESS writing example file ",
FILE_NAME, "! "
! MPI library must be shut down.
call MPI_Finalize(ierr)
contains
subroutine check(status)
integer, intent ( in) :: status
if(status /= nf90_noerr) then
print *, trim(nf90_strerror(status))
stop 2
end if
end subroutine check
end program simple_xy_par_wr
