Changes between Version 291 and Version 292 of doc/app/initialization_parameters

Timestamp:

Jun 10, 2016 12:31:18 PM (9 years ago)

Author:

suehring

Comment:

--

doc/app/initialization_parameters

@@ -806 +806 @@
 }}}
 {{{#!td
-Switch for topography boundary conditions in multigrid solver.\\\\
+Switch for topography boundary conditions in 'multigrid_noopt' solver.\\\\
 By default, Neumann boundary conditions for perturbation pressure are used in the multigrid solver at all wall boundaries. In case of '''masking_method''' = ''.TRUE.'', the masking method is used instead (i.e. the solver runs through the topography). \\\\
-'''Note: ''' If [#psolver psolver] = 'multigrid_fast', masking is always used, so that setting '''masking_method''' has no effect on the simulation at all.\\ \\
-'''Warning: ''' In case of complex topography and '''masking_method''' = ''.FALSE.'', the wind velocity may possibly blow up at certain grid points. Hence, so far it is recommended to use '''masking_method''' = ''.TRUE.''.  
+'''Note: ''' '''masking_method''' has only an effect if [#psolver psolver] = 'multigrid_noopt', otherwise, if [#psolver psolver] = 'multigrid', masking is always used, so that setting '''masking_method''' has no effect on the simulation at all.\\ \\
+'''Warning: ''' In case of complex topography and '''masking_method''' = ''.FALSE.'', the wind velocity may possibly blow up at certain grid points. Hence, so far it is recommended to use '''masking_method''' = ''.TRUE.'' in case [#psolver psolver] = 'multigrid_noopt'.  
 }}}
 |----------------
@@ -944 +944 @@
       This scheme requires that the number of grid points of the subdomains (or of the total domain, if only one PE is used) along each of the directions can, at least, be divided once by 2 without rest.\\\\
       With parallel runs, starting from a certain grid level the data of the subdomains are possibly gathered on PE0 in order to allow for a further coarsening of the grid. The grid level for gathering can be manually set by [#mg_switch_to_pe0_level mg_switch_to_pe0_level].\\\\
-      By default, Neumann boundary conditions for the perturbation pressure are used at all wall boundaries. In case of [#masking_method masking_method] = ''.TRUE.'', the masking method is used instead (i.e. the solver ''runs'' through the topography).\\\\
-'' 'multigrid_fast' ''\\\\
-      Same as '' 'multigrid' '' but optimized for better performance. Data along the {{{k}}} index (vertical direction) is sorted from alternate even and odd indices into two blocks of completely even and odd indices respectively, which allows vectorization on INTEL processors of loops in the red/black scheme. Additionally, topography flags are completely ignored (i.e. not used) in case of [#masking_method masking_method] switched on. On fine grid levels only even or odd data are send for the lateral ghost point exchange. Speedup of the multigrid solver for medium sized cases (1024*1024*192 grid points) can be about 40%.
+     'multigrid' is highly optimized for better performance. Data along the {{{k}}} index (vertical direction) is sorted from alternate even and odd indices into two blocks of completely even and odd indices respectively, which allows vectorization on INTEL processors of loops in the red/black scheme. Additionally, topography flags are completely ignored (i.e. not used) in case of [#masking_method masking_method] switched off. On fine grid levels only even or odd data are send for the lateral ghost point exchange. Speedup of the multigrid solver for medium sized cases (1024*1024*192 grid points) can be about 40% compared to 'multigrid_noopt'. \\
+     In contrast to 'multigrid_noopt', masking is always used, i.e. the solver runs through the topography similar to the FFT. 
+
+'' 'multigrid_noopt' ''\\\\
+    Similar as 'multigrid', but less optimized. However, in contrast to 'multigrid', 'multigrid_noopt' allows [#masking_method masking_method]=.FALSE., i.e. setting of Neumann boundary conditions for the perturbation pressure at all wall boundaries. \\\\
+
 '' 'sor' ''\\\\
       Successive over-relaxation method (SOR). The convergence of this iterative scheme is steered with the parameters [#omega_sor omega_sor], [#nsor_ini nsor_ini] and [#nsor nsor].\\