===== Domain_Features =====
This data set defines the domain size, the grid data, the domain decomposition features (MPI parallelisation characteristics : number of MPI processes connected to subdomains and how they are distributed over the domain) and the number of threads also used to split the domain (OpenMP parallelization).

===== Full data set of the namelist =====

   &Domain_Features Geometric_Layout               = 0,
                    Start_Coordinate_I_Direction   =-0.05 , 
                    End_Coordinate_I_Direction     = 0.05,
                    Start_Coordinate_J_Direction   =-0.05 , 
                    End_Coordinate_J_Direction     = 0.05,
                    Start_Coordinate_K_Direction   = 0.00 , 
                    End_Coordinate_K_Direction     = 0.00,
                    Cells_Number_I_Direction       = 80 ,
                    Cells_Number_J_Direction       = 60 ,
                    Cells_Number_K_Direction       = 1,
                    Number_OMP_Threads             = 1,
                    MPI_Cartesian_Topology         = .false. ,
                    MPI_Graphic_Topology           = .false. ,
                    Total_Number_MPI_Processes     = 1,
                    Max_Number_MPI_Proc_I_Direction= 1 , 
                    Max_Number_MPI_Proc_J_Direction= 1, 
                    Max_Number_MPI_Proc_K_Direction= 1,
                    Regular_Mesh                   = .true. /
           
{{ :sunfluidh:sketch_domain_2d.jpg?direct&600 |}}
           
<note important> 
In domain decomposition approach (MPI parallelization), the number of cells (Cells_Number_I_Direction, Cells_Number_J_Direction, Cells_Number_K_Direction) is related to each subdomain , __not the whole domain__.
Find some examples [[Domain_Features_Examples | here ]]. </note>
-----
===== Definition of the data set =====
-----
==== Geometric_Layout ====
   * Type : integer value
   * This option selects the type of geometry configuration used :
     * 0 : Cartesian geometry
       * 1: Cylindrical geometry. The axis is oriented along the K-direction. The coordinate system is $r(i), \theta(j) , z(k)$
       * 2: Cylindrical geometry. The axis is oriented along the I-direction. The coordinate system is $r(j), \theta(k) , z(i)$
       * 3: Cylindrical geometry. The axis is oriented along the J-direction. The coordinate system is $r(k),\theta(i),z(j)$  
       * Default value = 0             
==== Start_Coordinate_I_Direction === 
   * Type : real value  
   * Origin coordinate along the I-direction.
   * Default value must be set by the user
==== Start_Coordinate_J_Direction ====
   * Type : real value  
   * Origin coordinate along the J-direction.
   * Default value must be set by the user
==== Start_Coordinate_K_Direction ====
   * Type : real value  
   * Origin coordinate along the K-direction.
   * Default value must be set by the user
==== End_Coordinate_I_Direction === 
   * Type : real value  
   * End coordinate along the I-direction.
   * Default value must be set by the user
==== End_Coordinate_J_Direction ====
   * Type : real value  
   * End coordinate along the J-direction.
   * Default value must be set by the user
==== End_Coordinate_K_Direction ====
   * Type : real value  
   * End coordinate along the K-direction.
   * Default value must be set by the user
==== Cells_Number_I_Direction ==== 
   * Type : integer value 
   * Number of cells along the I-direction (excluding the ghost-cells).  \\ In decomposition domain approach (MPI parallelization), the number of cells is related to each subdomain , __not the whole domain__.
   * Default value= 0
==== Cells_Number_J_Direction ==== 
   * Type : integer value 
   * Number of cells along the J-direction (excluding the ghost-cells). \\ In decomposition domain approach (MPI parallelization), the number of cells is related to each subdomain , __not the whole domain__.
   * Default value= 0
==== Cells_Number_K_Direction ==== 
   * Type : integer value 
   * Number of cells along the K-direction (excluding the ghost-cells) \\ In decomposition domain approach (MPI parallelization), the number of cells is related to each subdomain , __not the whole domain__.
   * Default value= 0
==== Number_OMP_Threads ====
   * integer value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Number of Threads for OpenMP parallelization
   * Default value= 1
==== MPI_Cartesian_Topology ==== 
   * Type : Boolean value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Select the MPI cartesian topology for the domain decomposition method (same number of subdomains along a given direction)
   * Default value= .false.
==== MPI_Graphic_Topology ==== 
   * Type : Boolean value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Select the MPI graphic topology for the domain decomposition method (the number of subdomain along a given direction is variable)
   * Default value= .false.
==== Total_Number_MPI_Processes ==== 
   * Type : integer value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Total number of MPI processes used in the domain decomposition method
   * Default value= 1
==== Max_Number_MPI_Proc_I_Direction ==== 
   * Type : integer value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Number of MPI processes along the I-direction (maximum number for the graphic topology)
   * Default value= 1
==== Max_Number_MPI_Proc_J_Direction ==== 
   * Type : integer value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Number of MPI processes along the J-direction (maximum number for the graphic topology)
   * Default value= 1
==== Max_Number_MPI_Proc_K_Direction ==== 
   * Type : integer value (__** Not for the release SUNFLUIDH_EDU**__ ).
   * Number of MPI processes along the K-direction (maximum number for the graphic topology)
   * Default value= 1
==== Regular_Mesh ==== 
   * Type : boolean value
   * if .true., the mesh size is regular along each direction and the gris is directly built by the code. 
   * If .false., the grid is irregular and the cell distribution is read in the specific files maillx_xxxxx.d, mailly_xxxxx.d and maillz_xxxxx.d (xxxxx corresponds to the subdomain/MPI-process number if the MPI domain-decomposition is used). These files are created from the mesh builder named meshgen.x. 
   * Default value= .true.