[cig-commits] commit 1868 by bangerth to /var/svn/dealii/aspect

[dealii.demon at gmail.com]

Revision 1868

More on ways to make Aspect faster.

U   trunk/aspect/doc/manual/manual.tex


http://www.dealii.org/websvn/revision.php?repname=Aspect+Repository&path=%2F&rev=1868&peg=1868

Diff:
Modified: trunk/aspect/doc/manual/manual.tex
===================================================================
--- trunk/aspect/doc/manual/manual.tex	2013-09-03 21:57:54 UTC (rev 1867)
+++ trunk/aspect/doc/manual/manual.tex	2013-09-04 03:11:43 UTC (rev 1868)
@@ -2128,7 +2128,9 @@
 you choose these tolerances larger.
 The parameters you can adjust are all listed in
 Section~
ef{parameters:global} and are located at the top level of the input
-file.
+file. In particular, the parameters you want to look at are 	exttt{Linear
+solver tolerance}, 	exttt{Temperature solver tolerance} and
+	exttt{Composition solver tolerance}.
 \index[prmindex]{Composition solver tolerance}
 \index[prmindexfull]{Composition solver tolerance}
 \index[prmindex]{Linear solver tolerance}
@@ -2146,16 +2148,39 @@
 point where you will want to stop.
 
 
-\subsubsection{Switching off pressure normalization}
-
-
 \subsubsection{Using lower order of temperature/compositional discretization}
 
 
 \subsubsection{Limiting postprocessing}
 
 
+\subsubsection{Switching off pressure normalization}
+In most practically relevant cases, the Stokes equations
+\eqref{eq:stokes-1}--\eqref{eq:stokes-2} only determine the pressure up to a
+constant because only the pressure gradient appears in the equations, not the
+actual value of it. However, unlike this ``mathematical'' pressure, we have a
+very specific notion of the ``physical'' pressure: namely a well-defined
+quantity that at the surface of Earth equals the air pressure, which compared to
+the hydrostatic pressure inside Earth is essentially zero. 
 
+As a consequence, the default in spect{} is to normalize the computed
+``mathematical'' pressure in such a way that either the mean pressure at the
+surface is zero (where the geometry model describes where the ``surface'' is,
+see Section~
ef{sec:geometry-models}), or that the mean pressure in the domain
+is zero. This normalization is important if your model describes densities,
+viscosities and other quantities in dependence of the pressure -- because you
+almost certainly had the ``physical'' pressure in mind, not some unspecified
+``mathematical'' one. On the other hand, if you have a material model in which
+the pressure does not enter, then you don't need to normalize the pressure at
+all -- simply go with whatever the solver provides. In that case, you can switch
+off pressure normalization by looking at the 	exttt{Pressure normalization}
+parameter at the top level of the input file, see
+Section~
ef{parameters:global}.
+\index[prmindex]{Pressure normalization}
+\index[prmindexfull]{Pressure normalization}
+
+
+
 \section{Run-time input parameters}
 \label{sec:parameters}