/[MITgcm]/manual/s_phys_pkgs/text/exch2.tex
ViewVC logotype

Diff of /manual/s_phys_pkgs/text/exch2.tex

Parent Directory Parent Directory | Revision Log Revision Log | View Revision Graph Revision Graph | View Patch Patch

revision 1.12 by afe, Tue Mar 16 21:52:15 2004 UTC revision 1.26 by molod, Wed Jun 28 15:35:07 2006 UTC
# Line 10  Line 10 
10  %%    o automatically inserted at \section{Reference}  %%    o automatically inserted at \section{Reference}
11    
12    
13  \section{exch2: Extended Cubed Sphere \mbox{Topology}}  \subsection{exch2: Extended Cubed Sphere \mbox{Topology}}
14  \label{sec:exch2}  \label{sec:exch2}
15    
16    
17  \subsection{Introduction}  \subsubsection{Introduction}
18    
19  The \texttt{exch2} package extends the original cubed  The \texttt{exch2} package extends the original cubed sphere topology
20  sphere topology configuration to allow more flexible domain  configuration to allow more flexible domain decomposition and
21  decomposition and parallelization.  Cube faces (also called  parallelization.  Cube faces (also called subdomains) may be divided
22  subdomains) may be divided into any number of tiles that divide evenly  into any number of tiles that divide evenly into the grid point
23  into the grid point dimensions of the subdomain.  Furthermore, the  dimensions of the subdomain.  Furthermore, the tiles can run on
24  individual tiles may be run on separate processors in different  separate processors individually or in groups, which provides for
25  combinations, and whether exchanges between particular tiles occur  manual compile-time load balancing across a relatively arbitrary
26  between different processors is determined at runtime.  This  number of processors.
 flexibility provides for manual compile-time load balancing across a  
 relatively arbitrary number of processors. \\  
27    
28  The exchange parameters are declared in  The exchange parameters are declared in
29  \filelink{pkg/exch2/W2\_EXCH2\_TOPOLOGY.h}{pkg-exch2-W2_EXCH2_TOPOLOGY.h}  \filelink{pkg/exch2/W2\_EXCH2\_TOPOLOGY.h}{pkg-exch2-W2_EXCH2_TOPOLOGY.h}
# Line 34  and assigned in Line 32  and assigned in
32  validity of the cube topology depends on the \file{SIZE.h} file as  validity of the cube topology depends on the \file{SIZE.h} file as
33  detailed below.  The default files provided in the release configure a  detailed below.  The default files provided in the release configure a
34  cubed sphere topology of six tiles, one per subdomain, each with  cubed sphere topology of six tiles, one per subdomain, each with
35  32$\times$32 grid points, all running on a single processor.  Both  32$\times$32 grid points, with all tiles running on a single processor.  Both
36  files are generated by Matlab scripts in  files are generated by Matlab scripts in
37  \file{utils/exch2/matlab-topology-generator}; see Section  \file{utils/exch2/matlab-topology-generator}; see Section
38  \ref{sec:topogen} \sectiontitle{Generating Topology Files for exch2}  \ref{sec:topogen} \sectiontitle{Generating Topology Files for exch2}
# Line 43  of these files with alternate topologies Line 41  of these files with alternate topologies
41  \file{utils/exch2/code-mods} along with the appropriate \file{SIZE.h}  \file{utils/exch2/code-mods} along with the appropriate \file{SIZE.h}
42  file for single-processor execution.  file for single-processor execution.
43    
44  \subsection{Invoking exch2}  \subsubsection{Invoking exch2}
45    
46  To use exch2 with the cubed sphere, the following conditions must be  To use exch2 with the cubed sphere, the following conditions must be
47  met: \\  met:
48    
49  $\bullet$ The exch2 package is included when \file{genmake2} is run.  \begin{itemize}
50    The easiest way to do this is to add the line \code{exch2} to the  \item The exch2 package is included when \file{genmake2} is run.  The
51    \file{profile.conf} file -- see Section    easiest way to do this is to add the line \code{exch2} to the
52    \ref{sect:buildingCode} \sectiontitle{Building the code} for general    \file{packages.conf} file -- see Section \ref{sect:buildingCode}
53    details. \\    \sectiontitle{Building the code} for general
54      details.
55  $\bullet$ An example of \file{W2\_EXCH2\_TOPOLOGY.h} and  
56    \file{w2\_e2setup.F} must reside in a directory containing code  \item An example of \file{W2\_EXCH2\_TOPOLOGY.h} and
57    linked when \file{genmake2} runs.  The safest place to put these    \file{w2\_e2setup.F} must reside in a directory containing files
58    is the directory indicated in the \code{-mods=DIR} command line    symbolically linked by the \file{genmake2} script.  The safest place
59    modifier (typically \file{../code}), or the build directory.  The    to put these is the directory indicated in the \code{-mods=DIR}
60    default versions of these files reside in \file{pkg/exch2} and are    command line modifier (typically \file{../code}), or the build
61    linked automatically if no other versions exist elsewhere in the    directory.  The default versions of these files reside in
62    link path, but they should be left untouched to avoid breaking    \file{pkg/exch2} and are linked automatically if no other versions
63    configurations other than the one you intend to modify.\\    exist elsewhere in the build path, but they should be left untouched
64      to avoid breaking configurations other than the one you intend to
65  $\bullet$ Files containing grid parameters, named    modify.
66    \file{tile00$n$.mitgrid} where $n$=[1,6] (one per subdomain), must  
67    be in the working directory when the MITgcm executable is run.  \item Files containing grid parameters, named \file{tile00$n$.mitgrid}
68    These files are provided in the example experiments for cubed sphere    where $n$=\code{(1:6)} (one per subdomain), must be in the working
69    configurations with 32$\times$32 cube sides and are non-trivial to    directory when the MITgcm executable is run.  These files are
70    generate -- please contact MITgcm support if you want to generate    provided in the example experiments for cubed sphere configurations
71    files for other configurations. \\    with 32$\times$32 cube sides -- please contact MITgcm support if you
72      want to generate files for other configurations.
73  $\bullet$ As always when compiling MITgcm, the file \file{SIZE.h} must  
74    be placed where \file{genmake2} will find it.  In particular for the  \item As always when compiling MITgcm, the file \file{SIZE.h} must be
75    exch2, the domain decomposition specified in \file{SIZE.h} must    placed where \file{genmake2} will find it.  In particular for exch2,
76    correspond with the particular configuration's topology specified in    the domain decomposition specified in \file{SIZE.h} must correspond
77      with the particular configuration's topology specified in
78    \file{W2\_EXCH2\_TOPOLOGY.h} and \file{w2\_e2setup.F}.  Domain    \file{W2\_EXCH2\_TOPOLOGY.h} and \file{w2\_e2setup.F}.  Domain
79    decomposition issues particular to exch2 are addressed in Section    decomposition issues particular to exch2 are addressed in Section
80    \ref{sec:topogen} \sectiontitle{Generating Topology Files for exch2}    \ref{sec:topogen} \sectiontitle{Generating Topology Files for exch2}
81    and \ref{sec:exch2mpi} \sectiontitle{exch2, SIZE.h, and MPI}; a more    and \ref{sec:exch2mpi} \sectiontitle{exch2, SIZE.h, and
82    general background on the subject relevant to MITgcm is presented in      Multiprocessing}; a more general background on the subject
83    Section \ref{sect:specifying_a_decomposition}    relevant to MITgcm is presented in Section
84    \sectiontitle{Specifying a decomposition}.\\    \ref{sect:specifying_a_decomposition}
85      \sectiontitle{Specifying a decomposition}.
86    \end{itemize}
87    
88  As of the time of writing the following examples use exch2 and may be  At the time of this writing the following examples use exch2 and may
89  used for guidance:  be used for guidance:
90    
91  \begin{verbatim}  \begin{verbatim}
92  verification/adjust_nlfs.cs-32x32x1  verification/adjust_nlfs.cs-32x32x1
# Line 98  verification/hs94.cs-32x32x5 Line 99  verification/hs94.cs-32x32x5
99    
100    
101    
102  \subsection{Generating Topology Files for exch2}  \subsubsection{Generating Topology Files for exch2}
103  \label{sec:topogen}  \label{sec:topogen}
104    
105  Alternate cubed sphere topologies may be created using the Matlab  Alternate cubed sphere topologies may be created using the Matlab
# Line 108  m-file Line 109  m-file
109  from the Matlab prompt (there are no parameters to pass) generates  from the Matlab prompt (there are no parameters to pass) generates
110  exch2 topology files \file{W2\_EXCH2\_TOPOLOGY.h} and  exch2 topology files \file{W2\_EXCH2\_TOPOLOGY.h} and
111  \file{w2\_e2setup.F} in the working directory and displays a figure of  \file{w2\_e2setup.F} in the working directory and displays a figure of
112  the topology via Matlab.  The other m-files in the directory are  the topology via Matlab -- figures \ref{fig:6tile}, \ref{fig:12tile},
113  subroutines of \file{driver.m} and should not be run ``bare'' except  and \ref{fig:24tile} are examples of the generated diagrams.  The other
114    m-files in the directory are
115    subroutines called from \file{driver.m} and should not be run ``bare'' except
116  for development purposes. \\  for development purposes. \\
117    
118  The parameters that determine the dimensions and topology of the  The parameters that determine the dimensions and topology of the
119  generated configuration are \code{nr}, \code{nb}, \code{ng},  generated configuration are \code{nr}, \code{nb}, \code{ng},
120  \code{tnx} and \code{tny}, and all are assigned early in the script. \\  \code{tnx} and \code{tny}, and all are assigned early in the script. \\
121    
122  The first three determine the size of the subdomains and  The first three determine the height and width of the subdomains and
123  hence the size of the overall domain.  Each one determines the number  hence the size of the overall domain.  Each one determines the number
124  of grid points, and therefore the resolution, along the subdomain  of grid points, and therefore the resolution, along the subdomain
125  sides in a ``great circle'' around each axis of the cube.  At the time  sides in a ``great circle'' around each the three spatial axes of the cube.  At the time
126  of this writing MITgcm requires these three parameters to be equal,  of this writing MITgcm requires these three parameters to be equal,
127  but they provide for future releases  to accomodate different  but they provide for future releases  to accomodate different
128  resolutions around the axes to allow (for example) greater resolution  resolutions around the axes to allow subdomains with differing resolutions.\\
 around the equator.\\  
129    
130  The parameters \code{tnx} and \code{tny} determine the dimensions of  The parameters \code{tnx} and \code{tny} determine the width and height of
131  the tiles into which the subdomains are decomposed, and must evenly  the tiles into which the subdomains are decomposed, and must evenly
132  divide the integer assigned to \code{nr}, \code{nb} and \code{ng}.  divide the integer assigned to \code{nr}, \code{nb} and \code{ng}.
133  The result is a rectangular tiling of the subdomain.  Figure  The result is a rectangular tiling of the subdomain.  Figure
134  \ref{fig:24tile} shows one possible topology for a twenty-four tile  \ref{fig:24tile} shows one possible topology for a twenty-four-tile
135  cube, and figure \ref{fig:12tile} shows one for twelve tiles. \\  cube, and figure \ref{fig:12tile} shows one for twelve tiles. \\
136    
137  \begin{figure}  \begin{figure}
# Line 139  cube, and figure \ref{fig:12tile} shows Line 141  cube, and figure \ref{fig:12tile} shows
141   }   }
142  \end{center}  \end{center}
143    
144  \caption{Plot of cubed sphere topology with a 32$\times$192 domain  \caption{Plot of a cubed sphere topology with a 32$\times$192 domain
145  divided into six 32$\times$32 subdomains, each of which is divided into four tiles  divided into six 32$\times$32 subdomains, each of which is divided
146  (\code{tnx=16, tny=16}) for a total of twenty-four tiles.  into four tiles of width \code{tnx=16} and height \code{tny=16} for a
147  } \label{fig:24tile}  total of twenty-four tiles.  The colored borders of the subdomains
148    represent the parameters \code{nr} (red), \code{nb} (blue), and
149    \code{ng} (green).  } \label{fig:24tile}
150  \end{figure}  \end{figure}
151    
152  \begin{figure}  \begin{figure}
# Line 151  divided into six 32$\times$32 subdomains Line 155  divided into six 32$\times$32 subdomains
155    \includegraphics{part6/s12t_16x32.ps}    \includegraphics{part6/s12t_16x32.ps}
156   }   }
157  \end{center}  \end{center}
158  \caption{Plot of cubed sphere topology with a 32$\times$192 domain  \caption{Plot of a cubed sphere topology with a 32$\times$192 domain
159  divided into six 32$\times$32 subdomains of two tiles each  divided into six 32$\times$32 subdomains of two tiles each
160   (\code{tnx=16, tny=32}).   (\code{tnx=16, tny=32}).
161  } \label{fig:12tile}  } \label{fig:12tile}
162  \end{figure}  \end{figure}
163    
164    \begin{figure}
165    \begin{center}
166     \resizebox{4in}{!}{
167      \includegraphics{part6/s6t_32x32.ps}
168     }
169    \end{center}
170    \caption{Plot of a cubed sphere topology with a 32$\times$192 domain
171    divided into six 32$\times$32 subdomains with one tile each
172    (\code{tnx=32, tny=32}).  This is the default configuration.
173      }
174    \label{fig:6tile}
175    \end{figure}
176    
177    
178  Tiles can be selected from the topology to be omitted from being  Tiles can be selected from the topology to be omitted from being
179  allocated memory and processors.  This tuning is useful in ocean  allocated memory and processors.  This tuning is useful in ocean
180  modeling for omitting tiles that fall entirely on land.  The tiles  modeling for omitting tiles that fall entirely on land.  The tiles
# Line 167  by their tile number in the topology, se Line 185  by their tile number in the topology, se
185    
186    
187    
188  \subsection{exch2, SIZE.h, and multiprocessing}  \subsubsection{exch2, SIZE.h, and Multiprocessing}
189  \label{sec:exch2mpi}  \label{sec:exch2mpi}
190    
191  Once the topology configuration files are created, the Fortran  Once the topology configuration files are created, the Fortran
192  parameters in \file{SIZE.h} must be configured to match.  Section  \code{PARAMETER}s in \file{SIZE.h} must be configured to match.
193  \ref{sect:specifying_a_decomposition} \sectiontitle{Specifying a  Section \ref{sect:specifying_a_decomposition} \sectiontitle{Specifying
194  decomposition} provides a general description of domain decomposition    a decomposition} provides a general description of domain
195  within MITgcm and its relation to \file{SIZE.h}. The current section  decomposition within MITgcm and its relation to \file{SIZE.h}. The
196  specifies certain constraints the exch2 package imposes as well as  current section specifies constraints that the exch2 package imposes
197  describes how to enable parallel execution with MPI. \\  and describes how to enable parallel execution with MPI.
198    
199  As in the general case, the parameters \varlink{sNx}{sNx} and  As in the general case, the parameters \varlink{sNx}{sNx} and
200  \varlink{sNy}{sNy} define the size of the individual tiles, and so  \varlink{sNy}{sNy} define the size of the individual tiles, and so
201  must be assigned the same respective values as \code{tnx} and  must be assigned the same respective values as \code{tnx} and
202  \code{tny} in \file{driver.m}.\\  \code{tny} in \file{driver.m}.
203    
204  The halo width parameters \varlink{OLx}{OLx} and \varlink{OLy}{OLy}  The halo width parameters \varlink{OLx}{OLx} and \varlink{OLy}{OLy}
205  have no special bearing on exch2 and may be assigned as in the general  have no special bearing on exch2 and may be assigned as in the general
206  case. The same holds for \varlink{Nr}{Nr}, the number of vertical  case. The same holds for \varlink{Nr}{Nr}, the number of vertical
207  levels in the model.\\  levels in the model.
208    
209  The parameters \varlink{nSx}{nSx}, \varlink{nSy}{nSy},  The parameters \varlink{nSx}{nSx}, \varlink{nSy}{nSy},
210  \varlink{nPx}{nPx}, and \varlink{nPy}{nPy} relate to the number of  \varlink{nPx}{nPx}, and \varlink{nPy}{nPy} relate to the number of
211  tiles and how they are distributed on processors.  When using exch2,  tiles and how they are distributed on processors.  When using exch2,
212  the tiles are stored in single dimension, and so  the tiles are stored in the $x$ dimension, and so
213  \code{\varlink{nSy}{nSy}=1} in all cases.  Since the tiles as  \code{\varlink{nSy}{nSy}=1} in all cases.  Since the tiles as
214  configured by exch2 cannot be split up accross processors without  configured by exch2 cannot be split up accross processors without
215  regenerating the topology, \code{\varlink{nPy}{nPy}=1} as well. \\  regenerating the topology, \code{\varlink{nPy}{nPy}=1} as well.
216    
217  The number of tiles MITgcm allocates and how they are distributed  The number of tiles MITgcm allocates and how they are distributed
218  between processors depends on \varlink{nPx}{nPx} and  between processors depends on \varlink{nPx}{nPx} and
219  \varlink{nSx}{nSx}.  \varlink{nSx}{nSx} is the number of tiles per  \varlink{nSx}{nSx}.  \varlink{nSx}{nSx} is the number of tiles per
220  processor and \varlink{nPx}{nPx} the number of processors.  The total  processor and \varlink{nPx}{nPx} is the number of processors.  The
221  number of tiles in the topology minus those listed in  total number of tiles in the topology minus those listed in
222  \file{blanklist.txt} must equal \code{nSx*nPx}. \\  \file{blanklist.txt} must equal \code{nSx*nPx}.  Note that in order to
223    obtain maximum usage from a given number of processors in some cases,
224    this restriction might entail sharing a processor with a tile that
225    would otherwise be excluded because it is topographically outside of
226    the domain and therefore in \file{blanklist.txt}.  For example,
227    suppose you have five processors and a domain decomposition of
228    thirty-six tiles that allows you to exclude seven tiles.  To evenly
229    distribute the remaining twenty-nine tiles among five processors, you
230    would have to run one ``dummy'' tile to make an even six tiles per
231    processor.  Such dummy tiles are \emph{not} listed in
232    \file{blanklist.txt}.
233    
234  The following is an example of \file{SIZE.h} for the twelve-tile  The following is an example of \file{SIZE.h} for the twelve-tile
235  configuration illustrated in figure \ref{fig:12tile} running on  configuration illustrated in figure \ref{fig:12tile} running on one
236  one processor: \\  processor:
237    
238  \begin{verbatim}  \begin{verbatim}
239        PARAMETER (        PARAMETER (
# Line 222  one processor: \\ Line 250  one processor: \\
250       &           Nr  =   5)       &           Nr  =   5)
251  \end{verbatim}  \end{verbatim}
252    
253  The following is an example for the twentyfour-tile topology in figure  The following is an example for the twenty-four-tile topology in
254  \ref{fig:24tile} running on six processors:  figure \ref{fig:24tile} running on six processors:
255    
256  \begin{verbatim}  \begin{verbatim}
257        PARAMETER (        PARAMETER (
# Line 241  The following is an example for the twen Line 269  The following is an example for the twen
269  \end{verbatim}  \end{verbatim}
270    
271    
272    \subsubsection{Key Variables}
   
   
 \subsection{Key Variables}  
273    
274  The descriptions of the variables are divided up into scalars,  The descriptions of the variables are divided up into scalars,
275  one-dimensional arrays indexed to the tile number, and two and three  one-dimensional arrays indexed to the tile number, and two and
276  dimensional arrays indexed to tile number and neighboring tile.  This  three-dimensional arrays indexed to tile number and neighboring tile.
277  division reflects the functionality of these variables: The  This division reflects the functionality of these variables: The
278  scalars are common to every part of the topology, the tile-indexed  scalars are common to every part of the topology, the tile-indexed
279  arrays to individual tiles, and the arrays indexed by tile and  arrays to individual tiles, and the arrays indexed by tile and
280  neighbor to relationships between tiles and their neighbors. \\  neighbor to relationships between tiles and their neighbors. \\
281    
282  \subsubsection{Scalars}  Scalars:
283    
284  The number of tiles in a particular topology is set with the parameter  The number of tiles in a particular topology is set with the parameter
285  \code{NTILES}, and the maximum number of neighbors of any tiles by  \code{NTILES}, and the maximum number of neighbors of any tiles by
# Line 266  generated by \file{driver.m}.\\ Line 291  generated by \file{driver.m}.\\
291  The scalar parameters \varlink{exch2\_domain\_nxt}{exch2_domain_nxt}  The scalar parameters \varlink{exch2\_domain\_nxt}{exch2_domain_nxt}
292  and \varlink{exch2\_domain\_nyt}{exch2_domain_nyt} express the number  and \varlink{exch2\_domain\_nyt}{exch2_domain_nyt} express the number
293  of tiles in the $x$ and $y$ global indices.  For example, the default  of tiles in the $x$ and $y$ global indices.  For example, the default
294  setup of six tiles has \code{exch2\_domain\_nxt=6} and  setup of six tiles (Fig. \ref{fig:6tile}) has
295  \code{exch2\_domain\_nyt=1}.  A topology of twenty-four square tiles,  \code{exch2\_domain\_nxt=6} and \code{exch2\_domain\_nyt=1}.  A
296  four per subdomain (as in figure \ref{fig:24tile}), will have  topology of twenty-four square tiles, four per subdomain (as in figure
297  \code{exch2\_domain\_nxt=12} and \code{exch2\_domain\_nyt=2}.  Note  \ref{fig:24tile}), will have \code{exch2\_domain\_nxt=12} and
298  that these parameters express the tile layout to allow global data  \code{exch2\_domain\_nyt=2}.  Note that these parameters express the
299  files that are tile-layout-neutral and have no bearing on the internal  tile layout in order to allow global data files that are tile-layout-neutral.
300  storage of the arrays.  The tiles are internally stored in a range  They have no bearing on the internal storage of the arrays.  The tiles
301  from [1,\varlink{bi}{bi}] the $x$ axis and $y$ axis variable  are stored internally in a range from \code{\varlink{bi}{bi}=(1:NTILES)} in the
302  \varlink{bj}{bj} is generally ignored within the package. \\  $x$ axis, and the $y$ axis variable \varlink{bj}{bj} is assumed to
303    equal \code{1} throughout the package. \\
304  \subsubsection{Arrays Indexed to Tile Number}  
305    Arrays indexed to tile number:
306  The following arrays are of size \code{NTILES}, are indexed to the  
307  tile number, and the indices are omitted in their descriptions. \\  The following arrays are of length \code{NTILES} and are indexed to
308    the tile number, which is indicated in the diagrams with the notation
309    \textsf{t}$n$.  The indices are omitted in the descriptions. \\
310    
311  The arrays \varlink{exch2\_tnx}{exch2_tnx} and  The arrays \varlink{exch2\_tnx}{exch2_tnx} and
312  \varlink{exch2\_tny}{exch2_tny} express the $x$ and $y$ dimensions of  \varlink{exch2\_tny}{exch2_tny} express the $x$ and $y$ dimensions of
313  each tile.  At present for each tile \texttt{exch2\_tnx=sNx} and  each tile.  At present for each tile \texttt{exch2\_tnx=sNx} and
314  \texttt{exch2\_tny=sNy}, as assigned in \file{SIZE.h} and described in  \texttt{exch2\_tny=sNy}, as assigned in \file{SIZE.h} and described in
315  section \ref{sec:exch2mpi} \sectiontitle{exch2, SIZE.h, and  Section \ref{sec:exch2mpi} \sectiontitle{exch2, SIZE.h, and
316  multiprocessing}.  Future releases of MITgcm are to allow varying tile  Multiprocessing}.  Future releases of MITgcm may allow varying tile
317  sizes. \\  sizes. \\
318    
319  The location of the tiles' Cartesian origin within a subdomain are  The arrays \varlink{exch2\_tbasex}{exch2_tbasex} and
320  determined by the arrays \varlink{exch2\_tbasex}{exch2_tbasex} and  \varlink{exch2\_tbasey}{exch2_tbasey} determine the tiles'
321  \varlink{exch2\_tbasey}{exch2_tbasey}.  These variables are used to  Cartesian origin within a subdomain  
322  relate the location of the edges of different tiles to each other.  As  and locate the edges of different tiles relative to each other.  As
323  an example, in the default six-tile topology ??  each index in these  an example, in the default six-tile topology (Fig. \ref{fig:6tile})
324  arrays are set to \code{0}.  The twentyfour-tile case discussed above  each index in these arrays is set to \code{0} since a tile occupies
325  will have values of \code{0} or \code{16}, depending on the quadrant  its entire subdomain.  The twenty-four-tile case discussed above will
326  the tile falls within the subdomain.  The array  have values of \code{0} or \code{16}, depending on the quadrant of the
327  \varlink{exch2\_myFace}{exch2_myFace} contains the number of the  tile within the subdomain.  The elements of the arrays
328  subdomain of each tile, numbered \code{(1:6)} in the case of the  \varlink{exch2\_txglobalo}{exch2_txglobalo} and
329    \varlink{exch2\_txglobalo}{exch2_txglobalo} are similar to
330    \varlink{exch2\_tbasex}{exch2_tbasex} and
331    \varlink{exch2\_tbasey}{exch2_tbasey}, but locate the tile edges within the
332    global address space, similar to that used by global output and input
333    files. \\
334    
335    The array \varlink{exch2\_myFace}{exch2_myFace} contains the number of
336    the subdomain of each tile, in a range \code{(1:6)} in the case of the
337  standard cube topology and indicated by \textbf{\textsf{f}}$n$ in  standard cube topology and indicated by \textbf{\textsf{f}}$n$ in
338  figures \ref{fig:12tile}) and \ref{fig:24tile}). \\  figures \ref{fig:12tile} and
339    \ref{fig:24tile}. \varlink{exch2\_nNeighbours}{exch2_nNeighbours}
340    contains a count of the neighboring tiles each tile has, and sets
341    the bounds for looping over neighboring tiles.
342    \varlink{exch2\_tProc}{exch2_tProc} holds the process rank of each
343    tile, and is used in interprocess communication.  \\
344    
 The elements of the arrays \varlink{exch2\_txglobalo}{exch2_txglobalo}  
 and \varlink{exch2\_txglobalo}{exch2_txglobalo} are similar to  
 \varlink{exch2\_tbasex}{exch2_tbasex} and  
 \varlink{exch2\_tbasey}{exch2_tbasey}, but locate the tiles within the  
 global address space, similar to that used by global files. \\  
345    
346  The arrays \varlink{exch2\_isWedge}{exch2_isWedge},  The arrays \varlink{exch2\_isWedge}{exch2_isWedge},
347  \varlink{exch2\_isEedge}{exch2_isEedge},  \varlink{exch2\_isEedge}{exch2_isEedge},
348  \varlink{exch2\_isSedge}{exch2_isSedge}, and  \varlink{exch2\_isSedge}{exch2_isSedge}, and
349  \varlink{exch2\_isNedge}{exch2_isNedge} are set to \code{1} if the  \varlink{exch2\_isNedge}{exch2_isNedge} are set to \code{1} if the
350  indexed tile lies on the edge of a subdomain, \code{0} if not.  The  indexed tile lies on the edge of its subdomain, \code{0} if
351  values are used within the topology generator to determine the  not.  The values are used within the topology generator to determine
352  orientation of neighboring tiles, and to indicate whether a tile lies  the orientation of neighboring tiles, and to indicate whether a tile
353  on the corner of a subdomain.  The latter case requires special  lies on the corner of a subdomain.  The latter case requires special
354  exchange and numerical handling for the singularities at the eight  exchange and numerical handling for the singularities at the eight
355  corners of the cube.  \varlink{exch2\_nNeighbours}{exch2_nNeighbours}  corners of the cube. \\
356  contains a count of how many neighboring tiles each tile has, and is  
 used for setting bounds for looping over neighboring tiles.  
 \varlink{exch2\_tProc}{exch2_tProc} holds the process rank of each  
 tile, and is used in interprocess communication.  \\  
357    
358  \subsubsection{Arrays Indexed to Tile Number and Neighbor}  Arrays Indexed to Tile Number and Neighbor:
359    
360  The following arrays are all of size  The following arrays have vectors of length \code{MAX\_NEIGHBOURS} and
361  \code{MAX\_NEIGHBOURS}$\times$\code{NTILES} and describe the  \code{NTILES} and describe the orientations between the the tiles. \\
 orientations between the the tiles. \\  
362    
363  The array \code{exch2\_neighbourId(a,T)} holds the tile number  The array \code{exch2\_neighbourId(a,T)} holds the tile number
364  \code{Tn} for each of the tile number \code{T}'s neighboring tiles  \code{Tn} for each of the tile number \code{T}'s neighboring tiles
365  \code{a}.  The neighbor tiles are indexed \code{(1:MAX\_NEIGHBOURS)}  \code{a}.  The neighbor tiles are indexed
366  in the order right to left on the north then south edges, and then top  \code{(1:exch2\_nNeighbours(T))} in the order right to left on the
367  to bottom on the east and west edges.  Maybe throw in a fig here, eh?  north then south edges, and then top to bottom on the east then west
368  \\  edges.  \\
369    
370  The \code{exch2\_opposingSend\_record(a,T)} array holds the index   The \code{exch2\_opposingSend\_record(a,T)} array holds the
371  \code{b} in \texttt{exch2\_neighbourId(b,Tn)} that holds the tile  index \code{b} of the element in \texttt{exch2\_neighbourId(b,Tn)}
372  number \code{T}.  In other words,  that holds the tile number \code{T}, given
373    \code{Tn=exch2\_neighborId(a,T)}.  In other words,
374  \begin{verbatim}  \begin{verbatim}
375     exch2_neighbourId( exch2_opposingSend_record(a,T),     exch2_neighbourId( exch2_opposingSend_record(a,T),
376                        exch2_neighbourId(a,T) ) = T                        exch2_neighbourId(a,T) ) = T
377  \end{verbatim}  \end{verbatim}
378  This provides a back-reference from the neighbor tiles. \\  This provides a back-reference from the neighbor tiles. \\
379    
380  The arrays \varlink{exch2\_pi}{exch2_pi},  The arrays \varlink{exch2\_pi}{exch2_pi} and
381  \varlink{exch2\_pj}{exch2_pj}, \varlink{exch2\_oi}{exch2_oi},  \varlink{exch2\_pj}{exch2_pj} specify the transformations of indices
382  \varlink{exch2\_oj}{exch2_oj}, \varlink{exch2\_oi\_f}{exch2_oi_f}, and  in exchanges between the neighboring tiles.  These transformations are
383  \varlink{exch2\_oj\_f}{exch2_oj_f} specify the transformations in  necessary in exchanges between subdomains because a horizontal dimension
384  exchanges between the neighboring tiles.  The dimensions of  in one subdomain
385  \code{exch2\_pi(t,N,T)} and \code{exch2\_pj(t,N,T)} are the neighbor  may map to other horizonal dimension in an adjacent subdomain, and
386  ID \code{N} and the tile number \code{T} as explained above, plus a  may also have its indexing reversed. This swapping arises from the
387  vector of length 2 containing transformation factors \code{t}.  The  ``folding'' of two-dimensional arrays into a three-dimensional
388  first element of the transformation vector indicates the factor  cube. \\
389  \code{t} by which variables representing the same vector component of  
390  a tile \code{T} will be multiplied in exchanges with neighbor  The dimensions of \code{exch2\_pi(t,N,T)} and \code{exch2\_pj(t,N,T)}
391  \code{N}, and the second element indicates the transform to the  are the neighbor ID \code{N} and the tile number \code{T} as explained
392  variable in the other direction.  As an example,  above, plus a vector of length \code{2} containing transformation
393  \code{exch2\_pi(1,N,T)} holds the transform of the $i$ component of a  factors \code{t}.  The first element of the transformation vector
394  vector variable in tile \code{T} to the $i$ component of tile  holds the factor to multiply the index in the same dimension, and the
395  \code{T}'s neighbor \code{N}, and \code{exch2\_pi(2,N,T)} hold the  second element holds the the same for the orthogonal dimension.  To
396  component of neighbor \code{N}'s $j$ component. \\  clarify, \code{exch2\_pi(1,N,T)} holds the mapping of the $x$ axis
397    index of tile \code{T} to the $x$ axis of tile \code{T}'s neighbor
398    \code{N}, and \code{exch2\_pi(2,N,T)} holds the mapping of \code{T}'s
399    $x$ index to the neighbor \code{N}'s $y$ index. \\
400    
401  Under the current cube topology, one of the two elements of  One of the two elements of \code{exch2\_pi} or \code{exch2\_pj} for a
402  \code{exch2\_pi} or \code{exch2\_pj} for a given tile \code{T} and  given tile \code{T} and neighbor \code{N} will be \code{0}, reflecting
403  neighbor \code{N} will be \code{0}, reflecting the fact that the two  the fact that the two axes are orthogonal.  The other element will be
404  vector components are orthogonal.  The other element will be 1 or -1,  \code{1} or \code{-1}, depending on whether the axes are indexed in
405  depending on whether the components are indexed in the same or  the same or opposite directions.  For example, the transform vector of
406  opposite directions.  For example, the transform vector of the arrays  the arrays for all tile neighbors on the same subdomain will be
407  for all tile neighbors on the same subdomain will be \code{(1,0)},  \code{(1,0)}, since all tiles on the same subdomain are oriented
408  since all tiles on the same subdomain are oriented identically.  A  identically.  An axis that corresponds to the orthogonal dimension
409  vector direction that corresponds to the orthogonal dimension with the  with the same index direction in a particular tile-neighbor
410  same index direction in a particular tile-neighbor orientation will  orientation will have \code{(0,1)}.  Those with the opposite index
411  have \code{(0,1)}, whereas those in the opposite index direction will  direction will have \code{(0,-1)} in order to reverse the ordering. \\
 have \code{(0,-1)}.  This needs some diagrams.  
412    
413    The arrays \varlink{exch2\_oi}{exch2_oi},
414    \varlink{exch2\_oj}{exch2_oj}, \varlink{exch2\_oi\_f}{exch2_oi_f}, and
415    \varlink{exch2\_oj\_f}{exch2_oj_f} are indexed to tile number and
416    neighbor and specify the relative offset within the subdomain of the
417    array index of a variable going from a neighboring tile \code{N} to a
418    local tile \code{T}.  Consider \code{T=1} in the six-tile topology
419    (Fig. \ref{fig:6tile}), where
420    
 {\footnotesize  
421  \begin{verbatim}  \begin{verbatim}
422  C      exch2_pi          :: X index row of target to source permutation         exch2_oi(1,1)=33
423  C                        :: matrix for each neighbour entry.                     exch2_oi(2,1)=0
424  C      exch2_pj          :: Y index row of target to source permutation         exch2_oi(3,1)=32
425  C                        :: matrix for each neighbour entry.                     exch2_oi(4,1)=-32
 C      exch2_oi          :: X index element of target to source  
 C                        :: offset vector for cell-centered quantities    
 C                        :: of each neighbor entry.                      
 C      exch2_oj          :: Y index element of target to source  
 C                        :: offset vector for cell-centered quantities    
 C                        :: of each neighbor entry.                      
 C      exch2_oi_f        :: X index element of target to source  
 C                        :: offset vector for face quantities            
 C                        :: of each neighbor entry.                      
 C      exch2_oj_f        :: Y index element of target to source  
 C                        :: offset vector for face quantities            
 C                        :: of each neighbor entry.                      
426  \end{verbatim}  \end{verbatim}
 }  
427    
428    The simplest case is \code{exch2\_oi(2,1)}, the southern neighbor,
429    which is \code{Tn=6}.  The axes of \code{T} and \code{Tn} have the
430    same orientation and their $x$ axes have the same origin, and so an
431    exchange between the two requires no changes to the $x$ index.  For
432    the western neighbor (\code{Tn=5}), \code{code\_oi(3,1)=32} since the
433    \code{x=0} vector on \code{T} corresponds to the \code{y=32} vector on
434    \code{Tn}.  The eastern edge of \code{T} shows the reverse case
435    (\code{exch2\_oi(4,1)=-32)}), where \code{x=32} on \code{T} exchanges
436    with \code{x=0} on \code{Tn=2}. \\
437    
438     The most interesting case, where \code{exch2\_oi(1,1)=33} and
439    \code{Tn=3}, involves a reversal of indices.  As in every case, the
440    offset \code{exch2\_oi} is added to the original $x$ index of \code{T}
441    multiplied by the transformation factor \code{exch2\_pi(t,N,T)}.  Here
442    \code{exch2\_pi(1,1,1)=0} since the $x$ axis of \code{T} is orthogonal
443    to the $x$ axis of \code{Tn}.  \code{exch2\_pi(2,1,1)=-1} since the
444    $x$ axis of \code{T} corresponds to the $y$ axis of \code{Tn}, but the
445    index is reversed.  The result is that the index of the northern edge
446    of \code{T}, which runs \code{(1:32)}, is transformed to
447    \code{(-1:-32)}. \code{exch2\_oi(1,1)} is then added to this range to
448    get back \code{(32:1)} -- the index of the $y$ axis of \code{Tn}
449    relative to \code{T}.  This transformation may seem overly convoluted
450    for the six-tile case, but it is necessary to provide a general
451    solution for various topologies. \\
452    
453    
 \subsection{Key Routines}  
454    
455    Finally, \varlink{exch2\_itlo\_c}{exch2_itlo_c},
456    \varlink{exch2\_ithi\_c}{exch2_ithi_c},
457    \varlink{exch2\_jtlo\_c}{exch2_jtlo_c} and
458    \varlink{exch2\_jthi\_c}{exch2_jthi_c} hold the location and index
459    bounds of the edge segment of the neighbor tile \code{N}'s subdomain
460    that gets exchanged with the local tile \code{T}.  To take the example
461    of tile \code{T=2} in the twelve-tile topology
462    (Fig. \ref{fig:12tile}): \\
463    
464    \begin{verbatim}
465           exch2_itlo_c(4,2)=17
466           exch2_ithi_c(4,2)=17
467           exch2_jtlo_c(4,2)=0
468           exch2_jthi_c(4,2)=33
469    \end{verbatim}
470    
471    Here \code{N=4}, indicating the western neighbor, which is
472    \code{Tn=1}.  \code{Tn} resides on the same subdomain as \code{T}, so
473    the tiles have the same orientation and the same $x$ and $y$ axes.
474    The $x$ axis is orthogonal to the western edge and the tile is 16
475    points wide, so \code{exch2\_itlo\_c} and \code{exch2\_ithi\_c}
476    indicate the column beyond \code{Tn}'s eastern edge, in that tile's
477    halo region. Since the border of the tiles extends through the entire
478    height of the subdomain, the $y$ axis bounds \code{exch2\_jtlo\_c} to
479    \code{exch2\_jthi\_c} cover the height of \code{(1:32)}, plus 1 in
480    either direction to cover part of the halo. \\
481    
482    For the north edge of the same tile \code{T=2} where \code{N=1} and
483    the neighbor tile is \code{Tn=5}:
484    
485  \subsection{References}  \begin{verbatim}
486           exch2_itlo_c(1,2)=0
487           exch2_ithi_c(1,2)=0
488           exch2_jtlo_c(1,2)=0
489           exch2_jthi_c(1,2)=17
490    \end{verbatim}
491    
492    \code{T}'s northern edge is parallel to the $x$ axis, but since
493    \code{Tn}'s $y$ axis corresponds to \code{T}'s $x$ axis, \code{T}'s
494    northern edge exchanges with \code{Tn}'s western edge.  The western
495    edge of the tiles corresponds to the lower bound of the $x$ axis, so
496    \code{exch2\_itlo\_c} and \code{exch2\_ithi\_c} are \code{0}, in the
497    western halo region of \code{Tn}. The range of
498    \code{exch2\_jtlo\_c} and \code{exch2\_jthi\_c} correspond to the
499    width of \code{T}'s northern edge, expanded by one into the halo. \\
500    
501    
502    \subsubsection{Key Routines}
503    
504    Most of the subroutines particular to exch2 handle the exchanges
505    themselves and are of the same format as those described in
506    \ref{sect:cube_sphere_communication} \sectiontitle{Cube sphere
507    communication}.  Like the original routines, they are written as
508    templates which the local Makefile converts from \code{RX} into
509    \code{RL} and \code{RS} forms. \\
510    
511    The interfaces with the core model subroutines are
512    \code{EXCH\_UV\_XY\_RX}, \code{EXCH\_UV\_XYZ\_RX} and
513    \code{EXCH\_XY\_RX}.  They override the standard exchange routines
514    when \code{genmake2} is run with \code{exch2} option.  They in turn
515    call the local exch2 subroutines \code{EXCH2\_UV\_XY\_RX} and
516    \code{EXCH2\_UV\_XYZ\_RX} for two and three-dimensional vector
517    quantities, and \code{EXCH2\_XY\_RX} and \code{EXCH2\_XYZ\_RX} for two
518    and three-dimensional scalar quantities.  These subroutines set the
519    dimensions of the area to be exchanged, call \code{EXCH2\_RX1\_CUBE}
520    for scalars and \code{EXCH2\_RX2\_CUBE} for vectors, and then handle
521    the singularities at the cube corners. \\
522    
523    The separate scalar and vector forms of \code{EXCH2\_RX1\_CUBE} and
524    \code{EXCH2\_RX2\_CUBE} reflect that the vector-handling subroutine
525    needs to pass both the $u$ and $v$ components of the physical vectors.
526    This swapping arises from the topological folding discussed above, where the
527    $x$ and $y$ axes get swapped in some cases, and is not an
528    issue with the scalar case. These subroutines call
529    \code{EXCH2\_SEND\_RX1} and \code{EXCH2\_SEND\_RX2}, which do most of
530    the work using the variables discussed above. \\
531    
532    \subsubsection{Experiments and tutorials that use exch2}
533    \label{sec:pkg:exch2:experiments}
534    
535    \begin{itemize}
536    \item{Held Suarez tutorial, in tutorial\_held\_suarez\_cs verification directory,
537    described in section \ref{sect:eg-hs} }
538    \end{itemize}

Legend:
Removed from v.1.12  
changed lines
  Added in v.1.26

  ViewVC Help
Powered by ViewVC 1.1.22