Skip to content

GitLab

Commit c6095d92 authored by Praetorius, Simon's avatar Praetorius, Simon
Browse files
Options

parameter documentation

parent f261db05
Hide whitespace changes
Inline Side-by-side
Showing with 282 additions and 79 deletions
doc/csv2latex.php 0 → 100644
View file @ c6095d92
<?php
// script to convert a csv-like file to latex code
$latex = '\documentclass[10pt,a4paper]{article}'."\n"
.'\usepackage[a4paper, left=2cm, right=2cm, top=2cm, bottom=2cm,landscape]{geometry}'."\n"
.'\usepackage{amsmath}'."\n"
.'\usepackage{amsfonts}'."\n"
.'\usepackage{amssymb}'."\n"
.'\usepackage{array}'."\n"
.'\usepackage{longtable}'."\n"
.'\usepackage{textcomp}'."\n"
.'\author{IWR}'."\n"
.'\parindent0cm'."\n"
.'\parsep.5ex'."\n"
.'\newcommand{\basis}{{(name)->}}'."\n"
.'\begin{document}'."\n";
$filesToRead = array();
$dir = './parameters';
if ($handle = opendir($dir)) {
while (false !== ($file = readdir($handle))) {
if ($file != "." && $file != ".." && preg_match('/\.inc$/',$file)>0) {
$filesToRead[] = $dir.'/'.$file;
}
}
closedir($handle);
}
sort($filesToRead, SORT_STRING);
foreach($filesToRead as $file) {
$row = 1;
if (($handle = fopen($file, "r")) !== FALSE) {
$writeData = false;
if (($data = fgetcsv($handle, 1000, "|")) !== FALSE) {
if (strlen(trim($data[0]))>0)
$latex .= '\section{'.$data[0].'}'."\n";
$latex .= $data[1]."\n";
$latex .= '{'."\n";
$latex .= '\small'."\n";
if (count($data)>2)
$latex .= $data[2]."\n";
$latex .= '\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}'."\n"
.'\hline'."\n"
.'\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\\\'."\n"
.'\hline\hline'."\n\n".'\hline';
$writeData = true;
}
if ($writeData) {
while (($data = fgetcsv($handle, 1000, "|")) !== FALSE) {
$latex .= '\basis '.$data[0].' & \textbf{'.$data[1].'} & ['.$data[2].'] & '.$data[3].'\\\\ \hline'."\n";
}
$latex .= '\end{longtable}'."\n".'}'."\n\n";
}
fclose($handle);
}
}
$latex .= '\end{document}';
if (($handle = fopen('parameters.tex', "w")) !== FALSE) {
fwrite($handle, $latex);
}
?>
\ No newline at end of file
doc/parallel_parameters.tex
View file @ c6095d92
\documentclass[10pt,a4paper]{article}
\usepackage[a4paper, left=1.5cm, right=2cm, top=2cm, bottom=2cm]{geometry}
\usepackage[a4paper, left=2cm, right=2cm, top=2cm, bottom=2cm,landscape]{geometry}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{array}
\usepackage{longtable}
\usepackage{longtable}
\usepackage{textcomp}
\author{IWR}
\newcommand{\basis}{{[name]->}}
\parindent0cm
\parsep.5ex
\newcommand{\basis}{{(name)->}}
\begin{document}
\section{MeshDistributor}
The label \texttt{[name]} represents the string that is passed to the
constructor of MeshDistributor. {\small
\begin{longtable}[c]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline
\basis repartitioning & \textbf{bool} & [0] & En/disables
repartitioning of the macro mesh, when derivation of mean number of
DOFs exceeds a threshold value. \\\hline
\basis repartition ith change & \textbf{int} & [20] & Sets number of
mesh changes to wait before threshold check for repartitioning will
be performed.\\\hline
\basis partitioner & \textbf{enum}\{parmetis, zoltan, simple\} &
[parmetis] & Defines the external tool that performs the
partitioning of the, e.g.~by graph-partitioning. {\em simple} does
not change the initial partitioning, i.e., its a random distribution
of the macro elements to the processors.\\\hline
\basis box partitioning & \textbf{bool} & [0] & If the macro mesh is
globally refined from macro.stand.2d or macro.stand.3d, than the
partitioner may compute the partitioning not based on triangled or
tetrahedron, but on (composed) rectangles or boxes. Till now this is
implemented only for 3D and Zoltan partitioner. \\\hline
\basis log main rank & \textbf{bool} & [0] & If set to {\em true},
stdout output will be printed only by the main rank 0. Otherwise,
all ranks stdout output will be created. \\\hline
\basis pre refine & \textbf{int} & [-1] & If set to -1, the number
of pre refinements for the macro mesh will be calculated for the
given number of processors. This value can be overwritten by setting
a value $>= 0$. \\\hline
\basis output->serialization filename & \textbf{string} & [] & Name
of the parallel serialization file. If at least one stationary
problem is serialized, this parameter must be set. \\\hline
\basis input->serialization filename & \textbf{string} & [] & Name
of the parallel deserialization file. If at least one stationary
problem is deserialized, this parameter must be set. \\\hline
\basis debug output dir & \textbf{string} & [] & Path name where
debug data should be written to. \\\hline
\basis write parallel debug file & \textbf{bool} & [0] & If set to
{\em true}, the parallelization will create for each rank a file
with the name ``mpi-dbg-\{rank-no\}.dat''. This files contain
information about all DOF indices in ranks domain. They may be
useful for debugging or some postprocessing steps. \\\hline
\end{longtable}}
\section{AdaptInfo}
Parameters for AdaptInfo(\texttt{[name]},\texttt{nComponents})
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis start time & \textbf{double} & [0.0] & Initial time\\ \hline
\basis timestep & \textbf{double} & [0.0] & Time step size to be used\\ \hline
\basis end time & \textbf{double} & [1.0] & Final time\\ \hline
\basis max iteration & \textbf{int} & [-1] & maximal allowed number of iterations of the adaptive procedure; if maxIteration $\leq$ 0, no iteration bound is used\\ \hline
\basis max timestep iteration & \textbf{int} & [30] & Maximal number of iterations for choosing a timestep\\ \hline
\basis max time iteration & \textbf{int} & [30] & Maximal number of time iterations\\ \hline
\basis min timestep & \textbf{double} & [0.0] & Minimal step size\\ \hline
\basis max timestep & \textbf{double} & [1.0] & Maximal step size\\ \hline
\basis number of timesteps & \textbf{int} & [0] & Per default this value is 0 and not used. If it is set to a non-zero value, the computation of the stationary problem is done nTimesteps times with a fixed timestep.\\ \hline
\end{longtable}
}
Parameters for AdaptInfo for each component separately. \texttt{(name):=(name)[$i$]} for $i=0,\ldots,$\texttt{nComponents}$-1$
{
\small
\renewcommand{\basis}{{(name)[i]->}}
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis tolerance & \textbf{double} & [0.0] & Tolerance for the (absolute or relative) error\\ \hline
\basis time tolerance & \textbf{double} & [0.0] & Time tolerance.\\ \hline
\basis coarsen allowed & \textbf{int \{0,1\}} & [0] & true if coarsening is allowed, false otherwise.\\ \hline
\basis refinement allowed & \textbf{int \{0,1\}} & [1] & true if refinement is allowed, false otherwise.\\ \hline
\basis refine bisections & \textbf{int} & [1] & parameter to tell the marking strategy how many bisections should be performed when an element is marked for refinement; usually the value is 1 or DIM\\ \hline
\basis coarsen bisections & \textbf{int} & [1] & parameter to tell the marking strategy how many bisections should be undone when an element is marked for coarsening; usually the value is 1 or DIM\\ \hline
\basis sum factor & \textbf{double} & [1.0] & factors to combine max and integral time estimate\\ \hline
\basis max factor & \textbf{double} & [0.0] & factors to combine max and integral time estimate\\ \hline
\end{longtable}
}
\section{AdaptInstationary}
\texttt{(name)} ist the first argument of the constructor: AdaptInstationary(\texttt{(name)},\ldots)
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis strategy & \textbf{int} & [0] & Strategy for choosing one timestep: strategy 0: Explicit strategy, strategy 1: Implicit strategy.\\ \hline
\basis time delta 1 & \textbf{double} & [0.7071] & Parameter $\delta_1$ used in time step reduction\\ \hline
\basis time delta 2 & \textbf{double} & [1.4142] & Parameter $\delta_2$ used in time step enlargement\\ \hline
\basis info & \textbf{int} & [10] & Info level (from AdaptBase)\\ \hline
\basis break when stable & \textbf{int} & [0] & If this parameter is 1 and the instationary problem is stable, hence the number of solver iterations to solve the problem is zero, the adaption loop will stop.\\ \hline
\basis time adaptivity debug mode & \textbf{bool} & [0] & In debug mode, the adapt loop will print information about timestep decreasing and increasing.\\ \hline
\basis queue->runtime & \textbf{int} & [-1] & Runtime of the queue (of the servers batch system) in seconds. If the problem runs on a computer/server without a time limited queue, the value is -1.\\ \hline
\basis queue->serialization filename & \textbf{string} & [\_\_serialized\_problem.ser] & Name of the file used to automatically serialize the problem.\\ \hline
\end{longtable}
}
\section{AdaptStationary}
\texttt{(name)} ist the first argument of the constructor: AdaptStationary(\texttt{(name)},\ldots)
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis info & \textbf{int} & [10] & Info level (from AdaptBase)\\ \hline
\end{longtable}
}
\section{HL\_SignedDistTraverse}
Reinitialization class HL\_SignedDistTraverse gets the label \texttt{[name]} as first argument in the constructor. {\small
Reinitialization class HL\_SignedDistTraverse(\texttt{(name)},\ldots) with \texttt{(name)} as first argument in the constructor.
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis tolerance & \textbf{double} & [] & Tolerance for Hopf-Lax update iteration loop.\\ \hline
\basis maximal number of iteration steps & \textbf{int} & [] & Maximal number of mesh iterations for Hopf-Lax update.\\ \hline
\basis Gauss-Seidel iteration & \textbf{bool} & [] & Indicates whether Gauss-Seidel or Jacobi iteration is used. 0: Jacobi, 1: Gauss-Seidel\\ \hline
\basis infinity value & \textbf{double} & [] & Initialization value ``inifinity'' for non-boundary vertices. Must be $>$ 1000\\ \hline
\basis boundary initialization & \textbf{int} & [] & Define boundary initialization strategy. 0: BoundaryElementLevelSetDist, 1: BoundaryElementTopDist, 2: BoundaryElementEdgeDist, 3: BoundaryElementNormalDist\\ \hline
\end{longtable}
}
\section{Mesh}
Mesh(\texttt{(name)},$\cdot$)
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis macro file name & \textbf{string} & [] & Filename for the macrofile\\ \hline
\basis value file name & \textbf{string} & [] & Filename for value file. \\ \hline
\basis periodic file & \textbf{string} & [] & filename for periodic file\\ \hline
\basis check & \textbf{int} & [1] & Check the mesh structure\\ \hline
\basis preserve coarse dofs & \textbf{bool} & [0] & When an element is refined, not all dofs of the coarse element must be part of the new elements. An example are centered dofs when using higher lagrange basis functions. The midpoint dof of the parents element is not a dof of the both children elements. Therefore, the dof can be deleted. In some situation, e.g., when using multigrid techniques, it can be necessary to store this coarse dofs. Then this variable must be set to true. If false, the not required coarse dofs will be deleted.\\ \hline
\basis macro weights & \textbf{string} & [] & file of file that contains weight number for each element. The higher the weigh, the more this element will be refined and thus more processors share this element.\\ \hline
\basis global refinements & \textbf{int} & [0] & Number of global refinements\\ \hline
\end{longtable}
}
\section{Meshdistributor}
The label \texttt{(name)} represents the string that is passed to the constructor of MeshDistributor.
{
\small
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis repartitioning & \textbf{bool} & [0] & En/disables repartitioning of the macro mesh, when derivation of mean number of DOFs exceeds a threshold value.\\ \hline
\basis repartition ith change & \textbf{int} & [20] & Sets number of mesh changes to wait before threshold check for repartitioning will be performed.\\ \hline
\basis partitioner & \textbf{enum\{parmetis, zoltan, simple\}} & [parmetis] & Defines the external tool that performs the partitioning of the, e.g.~by graph-partitioning. {\em simple} does not change the initial partitioning, i.e., its a random distribution of the macro elements to the processors.\\ \hline
\basis box partitioning & \textbf{bool} & [0] & If the macro mesh is globally refined from macro.stand.2d or macro.stand.3d, than the partitioner may compute the partitioning not based on triangled or tetrahedron, but on (composed) rectangles or boxes. Till now this is implemented only for 3D and Zoltan partitioner.\\ \hline
\basis log main rank & \textbf{bool} & [0] & If set to {\em true}, stdout output will be printed only by the main rank 0. Otherwise, all ranks stdout output will be created.\\ \hline
\basis pre refine & \textbf{int} & [-1] & If set to -1, the number of pre refinements for the macro mesh will be calculated for the given number of processors. This value can be overwritten by setting a value $>= 0$.\\ \hline
\basis output->serialization filename & \textbf{string} & [] & Name of the parallel serialization file. If at least one stationary problem is serialized, this parameter must be set.\\ \hline
\basis input->serialization filename & \textbf{string} & [] & Name of the parallel deserialization file. If at least one stationary problem is deserialized, this parameter must be set.\\ \hline
\basis debug output dir & \textbf{string} & [] & Path name where debug data should be written to.\\ \hline
\basis write parallel debug file & \textbf{bool} & [0] & If set to {\em true}, the parallelization will create for each rank a file with the name ``mpi-dbg-\{rank-no\}.dat''. This files contain information about all DOF indices in ranks domain. They may be useful for debugging or some postprocessing steps.\\ \hline
\end{longtable}
}
\section{OEMSolver}
Standard parameters for OEMSolver
{
\small
\renewcommand{\basis}{{(solver)->}}
\begin{longtable}[l]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline\basis left precon & \textbf{enum \{diag,ilu,ic,no\}} & [no] & left preconditioner\\ \hline
\basis right precon & \textbf{enum \{diag,ilu,ic,no\}} & [no] & right preconditioner\\ \hline
\basis ell & \textbf{int} & [1] & additional solver parameter\\ \hline
\basis tolerance & \textbf{double} & [0.0] & Solver tolerance norm(r).\\ \hline
\basis relative tolerance & \textbf{double} & [0.0] & Relative solver tolerance norm(r)/norm(r0).\\ \hline
\basis max iteration & \textbf{int} & [1000] & maximal number of iterations.\\ \hline
\basis print cycle & \textbf{int} & [100] & Print cycle, after how many iterations the residuum norm is logged.\\ \hline
\basis info & \textbf{int} & [0] & info level during solving the system.\\ \hline
\end{longtable}
}
\begin{longtable}[c]{|>{\ttfamily}lp{.15\textwidth}>{\ttfamily}lp{.4\textwidth}|}
\hline
\textrm{\textbf{keyword}} & \textrm{\textbf{data type}} & \textrm{\textbf{default}} & \textrm{\textbf{description}} \\
\hline\hline
\hline
\basis tolerance & \textbf{double} & & Tolerance for Hopf-Lax update iteration loop.\\\hline
\basis maximal number of iteration steps & \textbf{int} & & Maximal number of mesh iterations for Hopf-Lax update.\\\hline
\basis Gauss-Seidel iteration & \textbf{bool} & & Indicates whether Gauss-Seidel or Jacobi iteration is used. 0: Jacobi, 1: Gauss-Seidel\\\hline
\basis infinity value & \textbf{double} & & Initialization value ``inifinity'' for non-boundary vertices. Must be $>$ 1000\\\hline
\basis boundary initialization & \textbf{int} & & Define boundary initialization strategy. 0: BoundaryElementLevelSetDist, 1: BoundaryElementTopDist, 2: BoundaryElementEdgeDist, 3: BoundaryElementNormalDist\\\hline
\end{longtable}}
\end{document}
\ No newline at end of file
doc/parameters/AdaptInfo.inc 0 → 100644
View file @ c6095d92
AdaptInfo|Parameters for AdaptInfo(\texttt{[name]},\texttt{nComponents})
start time|double|0.0|Initial time
timestep|double|0.0|Time step size to be used
end time|double|1.0|Final time
max iteration|int|-1|maximal allowed number of iterations of the adaptive procedure; if maxIteration $\leq$ 0, no iteration bound is used
max timestep iteration|int|30|Maximal number of iterations for choosing a timestep
max time iteration|int|30|Maximal number of time iterations
min timestep|double|0.0|Minimal step size
max timestep|double|1.0|Maximal step size
number of timesteps|int|0|Per default this value is 0 and not used. If it is set to a non-zero value, the computation of the stationary problem is done nTimesteps times with a fixed timestep.
\ No newline at end of file
doc/parameters/AdaptInfo_ScalContent.inc 0 → 100644
View file @ c6095d92
|Parameters for AdaptInfo for each component separately. \texttt{(name):=(name)[$i$]} for $i=0,\ldots,$\texttt{nComponents}$-1$|\renewcommand{\basis}{{(name)[i]->}}
tolerance|double|0.0|Tolerance for the (absolute or relative) error
time tolerance|double|0.0|Time tolerance.
coarsen allowed|int \{0,1\}|0|true if coarsening is allowed, false otherwise.
refinement allowed|int \{0,1\}|1|true if refinement is allowed, false otherwise.
refine bisections|int|1|parameter to tell the marking strategy how many bisections should be performed when an element is marked for refinement; usually the value is 1 or DIM
coarsen bisections|int|1|parameter to tell the marking strategy how many bisections should be undone when an element is marked for coarsening; usually the value is 1 or DIM
sum factor|double|1.0|factors to combine max and integral time estimate
max factor|double|0.0|factors to combine max and integral time estimate
\ No newline at end of file
doc/parameters/AdaptInstationary.inc 0 → 100644
View file @ c6095d92
AdaptInstationary|\texttt{(name)} ist the first argument of the constructor: AdaptInstationary(\texttt{(name)},\ldots)
strategy|int|0|Strategy for choosing one timestep: strategy 0: Explicit strategy, strategy 1: Implicit strategy.
time delta 1|double|0.7071|Parameter $\delta_1$ used in time step reduction
time delta 2|double|1.4142|Parameter $\delta_2$ used in time step enlargement
info|int|10|Info level (from AdaptBase)
break when stable|int|0|If this parameter is 1 and the instationary problem is stable, hence the number of solver iterations to solve the problem is zero, the adaption loop will stop.
time adaptivity debug mode|bool|0|In debug mode, the adapt loop will print information about timestep decreasing and increasing.
queue->runtime|int|-1|Runtime of the queue (of the servers batch system) in seconds. If the problem runs on a computer/server without a time limited queue, the value is -1.
queue->serialization filename|string|\_\_serialized\_problem.ser|Name of the file used to automatically serialize the problem.
\ No newline at end of file
doc/parameters/AdaptStationary.inc 0 → 100644
View file @ c6095d92
AdaptStationary|\texttt{(name)} ist the first argument of the constructor: AdaptStationary(\texttt{(name)},\ldots)
info|int|10|Info level (from AdaptBase)
\ No newline at end of file
doc/parameters/HL_SignedDistTraverse.inc 0 → 100644
View file @ c6095d92
HL\_SignedDistTraverse|Reinitialization class HL\_SignedDistTraverse(\texttt{(name)},\ldots) with \texttt{(name)} as first argument in the constructor.
tolerance|double||Tolerance for Hopf-Lax update iteration loop.
maximal number of iteration steps|int||Maximal number of mesh iterations for Hopf-Lax update.
Gauss-Seidel iteration|bool||Indicates whether Gauss-Seidel or Jacobi iteration is used. 0: Jacobi, 1: Gauss-Seidel
infinity value|double||Initialization value ``inifinity'' for non-boundary vertices. Must be $>$ 1000
boundary initialization|int||Define boundary initialization strategy. 0: BoundaryElementLevelSetDist, 1: BoundaryElementTopDist, 2: BoundaryElementEdgeDist, 3: BoundaryElementNormalDist
\ No newline at end of file
doc/parameters/Mesh.inc 0 → 100644
View file @ c6095d92
Mesh|Mesh(\texttt{(name)},$\cdot$)
macro file name|string||Filename for the macrofile
value file name|string||Filename for value file.
periodic file|string||filename for periodic file
check|int|1|Check the mesh structure
preserve coarse dofs|bool|0|When an element is refined, not all dofs of the coarse element must be part of the new elements. An example are centered dofs when using higher lagrange basis functions. The midpoint dof of the parents element is not a dof of the both children elements. Therefore, the dof can be deleted. In some situation, e.g., when using multigrid techniques, it can be necessary to store this coarse dofs. Then this variable must be set to true. If false, the not required coarse dofs will be deleted.
macro weights|string||file of file that contains weight number for each element. The higher the weigh, the more this element will be refined and thus more processors share this element.
global refinements|int|0|Number of global refinements
\ No newline at end of file
doc/parameters/MeshDistributor.inc 0 → 100644
View file @ c6095d92
Meshdistributor|The label \texttt{(name)} represents the string that is passed to the constructor of MeshDistributor.
repartitioning|bool|0|En/disables repartitioning of the macro mesh, when derivation of mean number of DOFs exceeds a threshold value.
repartition ith change|int|20|Sets number of mesh changes to wait before threshold check for repartitioning will be performed.
partitioner|enum\{parmetis, zoltan, simple\}|parmetis|Defines the external tool that performs the partitioning of the, e.g.~by graph-partitioning. {\em simple} does not change the initial partitioning, i.e., its a random distribution of the macro elements to the processors.
box partitioning|bool|0|If the macro mesh is globally refined from macro.stand.2d or macro.stand.3d, than the partitioner may compute the partitioning not based on triangled or tetrahedron, but on (composed) rectangles or boxes. Till now this is implemented only for 3D and Zoltan partitioner.
log main rank|bool|0|If set to {\em true}, stdout output will be printed only by the main rank 0. Otherwise, all ranks stdout output will be created.
pre refine|int|-1|If set to -1, the number of pre refinements for the macro mesh will be calculated for the given number of processors. This value can be overwritten by setting a value $>= 0$.
output->serialization filename|string||Name of the parallel serialization file. If at least one stationary problem is serialized, this parameter must be set.
input->serialization filename|string||Name of the parallel deserialization file. If at least one stationary problem is deserialized, this parameter must be set.
debug output dir|string||Path name where debug data should be written to.
write parallel debug file|bool|0|If set to {\em true}, the parallelization will create for each rank a file with the name ``mpi-dbg-\{rank-no\}.dat''. This files contain information about all DOF indices in ranks domain. They may be useful for debugging or some postprocessing steps.
\ No newline at end of file
doc/parameters/OEMSolver.inc 0 → 100644
View file @ c6095d92
OEMSolver|Standard parameters for OEMSolver|\renewcommand{\basis}{{(solver)->}}
left precon|enum \{diag,ilu,ic,no\}|no|left preconditioner
right precon|enum \{diag,ilu,ic,no\}|no|right preconditioner
ell|int|1|additional solver parameter
tolerance|double|0.0|Solver tolerance norm(r).
relative tolerance|double|0.0|Relative solver tolerance norm(r)/norm(r0).
max iteration|int|1000|maximal number of iterations.
print cycle|int|100|Print cycle, after how many iterations the residuum norm is logged.
info|int|0|info level during solving the system.
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment