Différences

Ci-dessous, les différences entre deux révisions de la page.

--- sunfluidh:domain_features_examples [2016/11/29 11:33] – yann
+++ sunfluidh:domain_features_examples [2017/09/26 17:15] (Version actuelle) – yann
@@ Ligne 77: / Ligne 77: @@
 <note important> The code must be compiled with the MPI options. </note>
-=== Parallel setting : MPI Only (in MPI graphic topology) ===
+=== Parallel setting : MPI Only (in MPI graph topology) ===
 <note> Domain in cartesian geometry.\\
 No OpenMP parallelization is considered .\\
-Domain decomposition approach (MPI parallelization) in MPI graphic topology. The domain is divided on 4 subdomains :
+Domain decomposition approach (MPI parallelization) in MPI graph topology. The domain is divided on 4 subdomains :
    * 4 along the I-direction (maximum value)
    * 2 along the J-direction (maximum value)
@@ Ligne 104: / Ligne 104: @@
 <WRAP important> The code must be compiled with the MPI options.\\
-The MPI-graphic topology is used in cases where the domain configuration have got large immersed bodies. \\ The aim is to build a domain decomposition in a way that takes the solid parts out of the domain and ensure that the numerical domain is mainly fluid.\\
+The MPI graph topology is used in cases where the domain configuration have got large immersed bodies. \\ The aim is to build a domain decomposition in a way that takes the solid parts out of the domain and ensure that the numerical domain is mainly fluid.\\
 In a first step, the domain decomposition is carried out as if the MPI cartesian decomposition was used. The number of processes "Total_Number_MPI_Processes" will be equal to the multiplication of the "Max_Number_MPI_Proc_I_Direction" by "Max_Number_MPI_Proc_J_Direction" by "Max_Number_MPI_Proc_K_Direction". \\
-When some subdomains are totally occupied by solid parts, they are useless. They must therefore be removed in order to reduce the MPI process number. As the MPI topology is no longer cartesian due to the "holes" arisen in the MPI topology, the subdomain decomposition is handled with the MPI graphic topology. \\
+When some subdomains are totally occupied by solid parts, they are useless. They must therefore be removed in order to reduce the MPI process number. As the MPI topology is no longer cartesian due to the "holes" arisen in the MPI topology, the subdomain decomposition is handled with the MPI graph topology. \\
 The software "mpi_subdomain_decomposition" has been developed for helping the user to build the subdomain decomposition in these cases :\\
@@ Ligne 113: / Ligne 113: @@
      *  reads the data file
      *  analyses the geometrical layout of the domain with respect to the subdomain decomposition following a MPI cartesian topology proposed by the user
-     * reformulates the subdomain splitting by excluding the useless subdomains (only bounded to solid parts of the domain).
+     * reformulates the subdomain splitting by excluding the pointless subdomains (covering the solid parts of the domain).
      * build the data file named "data_mpi_subdomain_layout.dat" that contains :
        * a map of the enabled MPI processes related to the new subdomain layout
-       * the maximum MPI process number implied in the MPI graphic topology
+       * the maximum MPI process number implied in the MPI graph topology
-The user must set the variable "Total_Number_MPI_Processes" to this new value and keep the other data in the same state. During the run, sunfluidh will read this own data file and the new data file "data_mpi_subdomain_layout.dat" in order to know the MPI process layout related to the MPI graphic topology.
+The user must set the variable "Total_Number_MPI_Processes" to this new value and keep the other data in the same state. During the run, sunfluidh will read this own data file and the new data file "data_mpi_subdomain_layout.dat" in order to know the MPI process layout related to the MPI graph topology.
  </WRAP>
 [[sunfluidh:domain_features_namelist|Click here to come back to the previous page]] \\