/[MITgcm]/manual/s_software/text/sarch.tex
ViewVC logotype

Diff of /manual/s_software/text/sarch.tex

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

revision 1.13 by afe, Wed Jan 28 20:27:54 2004 UTC revision 1.19 by edhill, Tue Apr 20 23:32:59 2004 UTC
# Line 76  floating point operations.} Line 76  floating point operations.}
76  \end{figure}  \end{figure}
77    
78  \section{WRAPPER}  \section{WRAPPER}
79    \begin{rawhtml}
80    <!-- CMIREDIR:wrapper: -->
81    \end{rawhtml}
82    
83  A significant element of the software architecture utilized in  A significant element of the software architecture utilized in
84  MITgcm is a software superstructure and substructure collectively  MITgcm is a software superstructure and substructure collectively
# Line 150  easily be specialized to fit, in a compu Line 153  easily be specialized to fit, in a compu
153  computer architecture currently available to the scientific computing community.  computer architecture currently available to the scientific computing community.
154    
155  \subsection{Machine model parallelism}  \subsection{Machine model parallelism}
156    \begin{rawhtml}
157    <!-- CMIREDIR:domain_decomp: -->
158    \end{rawhtml}
159    
160   Codes operating under the WRAPPER target an abstract machine that is assumed to   Codes operating under the WRAPPER target an abstract machine that is assumed to
161  consist of one or more logical processors that can compute concurrently.    consist of one or more logical processors that can compute concurrently.  
# Line 977  program, MPI is used. The calls to the M Line 983  program, MPI is used. The calls to the M
983  routines must be activated at compile time.  Currently MPI libraries are  routines must be activated at compile time.  Currently MPI libraries are
984  invoked by  invoked by
985  specifying the appropriate options file with the  specifying the appropriate options file with the
986  \begin{verbatim}-of\end{verbatim} flag when running the {\em genmake2}  {\tt-of} flag when running the {\em genmake2}
987  script, which generates the Makefile for compiling and linking MITgcm.  script, which generates the Makefile for compiling and linking MITgcm.
988  (Previously this was done by setting the {\em ALLOW\_USE\_MPI} and  (Previously this was done by setting the {\em ALLOW\_USE\_MPI} and
989  {\em ALWAYS\_USE\_MPI} flags in the {\em CPP\_EEOPTIONS.h} file.)  More  {\em ALWAYS\_USE\_MPI} flags in the {\em CPP\_EEOPTIONS.h} file.)  More
990  detailed information about the use of {\em genmake2} for specifying  detailed information about the use of {\em genmake2} for specifying
991  local compiler flags is located in section 3 ??\\    local compiler flags is located in section \ref{sect:genmake}.\\  
992    
993    
994  \fbox{  \fbox{
995  \begin{minipage}{4.75in}  \begin{minipage}{4.75in}
996    Directory: {\em tools/build\_options}\\
997  File: {\em tools/genmake2}  File: {\em tools/genmake2}
998  \end{minipage}  \end{minipage}
999  } \\  } \\
# Line 1005  product of the processor grid settings o Line 1012  product of the processor grid settings o
1012  in the file {\em SIZE.h}. The parameter {\em mf} specifies that a text file  in the file {\em SIZE.h}. The parameter {\em mf} specifies that a text file
1013  called ``mf'' will be read to get a list of processor names on  called ``mf'' will be read to get a list of processor names on
1014  which the sixty-four processes will execute. The syntax of this file  which the sixty-four processes will execute. The syntax of this file
1015  is specified by the MPI distribution  is specified by the MPI distribution.
1016  \\  \\
1017    
1018  \fbox{  \fbox{
# Line 1056  to processor identification are are used Line 1063  to processor identification are are used
1063  Allocation of processes to tiles in controlled by the routine  Allocation of processes to tiles in controlled by the routine
1064  {\em INI\_PROCS()}. For each process this routine sets  {\em INI\_PROCS()}. For each process this routine sets
1065  the variables {\em myXGlobalLo} and {\em myYGlobalLo}.  the variables {\em myXGlobalLo} and {\em myYGlobalLo}.
1066  These variables specify (in index space) the coordinate  These variables specify in index space the coordinates
1067  of the southern most and western most corner of the  of the southernmost and westernmost corner of the
1068  southern most and western most tile owned by this process.  southernmost and westernmost tile owned by this process.
1069  The variables {\em pidW}, {\em pidE}, {\em pidS} and {\em pidN}  The variables {\em pidW}, {\em pidE}, {\em pidS} and {\em pidN}
1070  are also set in this routine. These are used to identify  are also set in this routine. These are used to identify
1071  processes holding tiles to the west, east, south and north  processes holding tiles to the west, east, south and north
# Line 1067  in the header file {\em EESUPPORT.h} for Line 1074  in the header file {\em EESUPPORT.h} for
1074  communication routines.  The above does not hold when the  communication routines.  The above does not hold when the
1075  exch2 package is used -- exch2 sets its own parameters to  exch2 package is used -- exch2 sets its own parameters to
1076  specify the global indices of tiles and their relationships  specify the global indices of tiles and their relationships
1077  to each other.  to each other.  See the documentation on the exch2 package
1078    (\ref{sec:exch2})  for
1079    details.
1080  \\  \\
1081    
1082  \fbox{  \fbox{
# Line 1093  operations and that can be customized fo Line 1102  operations and that can be customized fo
1102  describes the information that is held and used.  describes the information that is held and used.
1103    
1104  \begin{enumerate}  \begin{enumerate}
1105  \item {\bf Tile-tile connectivity information} For each tile the WRAPPER  \item {\bf Tile-tile connectivity information}
1106  sets a flag that sets the tile number to the north, south, east and  For each tile the WRAPPER
1107    sets a flag that sets the tile number to the north,
1108    south, east and
1109  west of that tile. This number is unique over all tiles in a  west of that tile. This number is unique over all tiles in a
1110  configuration. The number is held in the variables {\em tileNo}  configuration. Except when using the cubed sphere and the exch2 package,
1111    the number is held in the variables {\em tileNo}
1112  ( this holds the tiles own number), {\em tileNoN}, {\em tileNoS},  ( this holds the tiles own number), {\em tileNoN}, {\em tileNoS},
1113  {\em tileNoE} and {\em tileNoW}. A parameter is also stored with each tile  {\em tileNoE} and {\em tileNoW}. A parameter is also stored with each tile
1114  that specifies the type of communication that is used between tiles.  that specifies the type of communication that is used between tiles.
# Line 1119  of the WRAPPER exchange primitive Line 1131  of the WRAPPER exchange primitive
1131  (see figure \ref{fig:communication_primitives}). The routine  (see figure \ref{fig:communication_primitives}). The routine
1132  {\em ini\_communication\_patterns()} is responsible for setting the  {\em ini\_communication\_patterns()} is responsible for setting the
1133  communication mode values for each tile.  communication mode values for each tile.
1134  \\  
1135    When using the cubed sphere configuration with the exch2 package, the
1136    relationships between tiles and their communication methods are set
1137    by the package in other variables.  See the exch2 package documentation
1138    (\ref{sec:exch2} for details.
1139    
1140    
1141    
1142  \fbox{  \fbox{
1143  \begin{minipage}{4.75in}  \begin{minipage}{4.75in}
# Line 1255  The \_EXCH operation is also central to Line 1273  The \_EXCH operation is also central to
1273  the cube-sphere grid. In this class of grid a rotation may be required  the cube-sphere grid. In this class of grid a rotation may be required
1274  between tiles. Aligning the coordinate requiring rotation with the  between tiles. Aligning the coordinate requiring rotation with the
1275  tile decomposition, allows the coordinate transformation to  tile decomposition, allows the coordinate transformation to
1276  be embedded within a custom form of the \_EXCH primitive.  be embedded within a custom form of the \_EXCH primitive.  In these
1277    cases \_EXCH is mapped to exch2 routines, as detailed in the exch2
1278    package documentation  \ref{sec:exch2}.
1279    
1280  \item {\bf Reverse Mode}  \item {\bf Reverse Mode}
1281  The communication primitives \_EXCH and \_GSUM both employ  The communication primitives \_EXCH and \_GSUM both employ
# Line 1272  operations. However, the routine argumen Line 1292  operations. However, the routine argumen
1292  is set to the value {\em REVERSE\_SIMULATION}. This signifies  is set to the value {\em REVERSE\_SIMULATION}. This signifies
1293  ti the low-level routines that the adjoint forms of the  ti the low-level routines that the adjoint forms of the
1294  appropriate communication operation should be performed.  appropriate communication operation should be performed.
1295    
1296  \item {\bf MAX\_NO\_THREADS}  \item {\bf MAX\_NO\_THREADS}
1297  The variable {\em MAX\_NO\_THREADS} is used to indicate the  The variable {\em MAX\_NO\_THREADS} is used to indicate the
1298  maximum number of OS threads that a code will use. This  maximum number of OS threads that a code will use. This
# Line 1376  a series of template files, for example Line 1397  a series of template files, for example
1397  This is done to allow a large number of variations on the exchange  This is done to allow a large number of variations on the exchange
1398  process to be maintained. One set of variations supports the  process to be maintained. One set of variations supports the
1399  cube sphere grid. Support for a cube sphere grid in MITgcm is based  cube sphere grid. Support for a cube sphere grid in MITgcm is based
1400  on having each face of the cube as a separate tile (or tiles).  on having each face of the cube as a separate tile or tiles.
1401  The exchange routines are then able to absorb much of the  The exchange routines are then able to absorb much of the
1402  detailed rotation and reorientation required when moving around the  detailed rotation and reorientation required when moving around the
1403  cube grid. The set of {\em \_EXCH} routines that contain the  cube grid. The set of {\em \_EXCH} routines that contain the
# Line 1400  quantities at the C-grid vorticity point Line 1421  quantities at the C-grid vorticity point
1421    
1422    
1423  \section{MITgcm execution under WRAPPER}  \section{MITgcm execution under WRAPPER}
1424    \begin{rawhtml}
1425    <!-- CMIREDIR:mitgcm_wrapper: -->
1426    \end{rawhtml}
1427    
1428  Fitting together the WRAPPER elements, package elements and  Fitting together the WRAPPER elements, package elements and
1429  MITgcm core equation elements of the source code produces calling  MITgcm core equation elements of the source code produces calling

Legend:
Removed from v.1.13  
changed lines
  Added in v.1.19

  ViewVC Help
Powered by ViewVC 1.1.22