io.yml

pdi:
  metadata: # type of small values for which PDI keeps a copy

     nx: int #Domain size per proc
     ny: int #Domain size per proc
     mpi_max_coords_x: int #MPI decomposition
     mpi_max_coords_y: int #MPI decomposition
     mpi_coords_x: int #MPI coordinate of the current process 
     mpi_coords_y: int #MPI coordinate of the current process 
     time: double #Time
     nwrite: int #Index for naming files


  data: # type of values for which PDI does not keep a copy
    main_field: { size: [ '$nx+2', '$ny+2' ], type: array, subtype: double } #Field of the current subdmain
  
  plugins:
    mpi:
    decl_hdf5: #Writing in hdf5
      file: data_${nwrite:04}.h5 #Name of the file
      communicator: $MPI_COMM_WORLD # the MPI communicator used for HDF5 parallel synchronized write
      datasets: #Dataset we are writing in: the whole solution
        data:
          type: array
          subtype: double
          size: ['$nx*$mpi_max_coords_x', '$ny*$mpi_max_coords_y']
      write:
        main_field: # the name of the data to write
          dataset: data
          memory_selection:  #exclude ghosts from the data in memory
            size:  ['$nx', '$ny']
            start: [1, 1]
          dataset_selection: #Write in the file at the current MPI process's coordinates
            size:  ['$nx', '$ny']
            start: ['$nx*$mpi_coords_x', '$ny*$mpi_coords_y']

        time: