Skip to content

Compilation and Execution

make

The directories src/swpc_3d, src/swpc_psv, src/swpc_sh, and src/tools contain makefile. Execute the make command in each directory to generate the executable binaries. An executable file (with a *.x extension) will be stored in the bin directory.

Specifying Compiler Options

In the makefiles, the following variables must be specified according to the environment:

variable description
FC compiler name
FFLAGS compiler option
NCFLAG NetCDF flag
NCLIB location of the NetCDF library directory
NCINC location of the NetCDF header file directory
NETCDF linker option for NetCDF

If NCFLAG = -D_NETCDF is specified, the make command will try to compile OpenSWPC with NetCDF library.

A set of the above variables under different computer environments is defined in src/shared/makefile.arch and src/shared/makefile-tools.arch. The former is for the compilation of FDM codes, and the latter is for the compilation of misc tools. The user can specify the arch option in the make command as in the following example: 

1
make arch=mac-gfortran

In the above case, the appropriate compiler options for the architecture mac-gfortran is automatically set. In some environments, the debug option is predefined. This option is used in the following way: 

1
make arch=eic debug=true

The list of pre-defined architecture (arch) options is described in the following table.

arch name target NetCDF location
mac-intel Mac OSX + Intel Compiler + OpenMPI ${HOME}/local
mac-gfortran Mac OSX + gfortran + Open MPI /usr/local
eic EIC (ERI, UTokyo) with the Intel Compiler ${HOME}/local
fx Fujitsu FX10, FX100 and the K-computer ${HOME}/xlocal
es3 The Earth Simulator 3 (obsolete) Provided by the system
es4 The Earth Simulator 4 Provided by the system
ubuntu-gfortran Ubuntu 16.04LTS + gfortran + Open MPI Installation by apt
ofp (or oak) Oakforest-PACS of the University of Tokyo (obsolete) automatically specified by the module command
obcx Oakbridge-CX of the University of Tokyo (obsolete) automatically specified by the module command
bdec-o Wisteria/BDEC-01 (Odyssey) of the University of Tokyo automatically specified by the module command
mac-m1 macOS + gfortran (Apple Silicon (M1/M2) + Homebrew) /opt/homebrew/

More about the NetCDF library

The NetCDF library consists of the following items:

  • libnetcdf.*: NetCDF library file
  • libnetcdff.*: NetCDF Fortran library file (only for the NetCDF version 4 or later)
  • netcdf.mod: Fortran module information file

The extension of the library files may be *.a (static library) or *.so (dynamic library), depending on the installation. All these files are necessary for successful compilation with NetCDF. In particular, the netcdf.mod file must be created by the same Fortran compiler as OpenSWPC. If NetCDF is installed using packaging tools such as yum, apt, or homebrew, the use of gfortran is implicitly assumed.

Parameters embedded in the source code

Although most of the controlling parameter of the OpenSWPC are given in the input parameter file at the time of execution, a few parameters shown below are embedded in the source code for improving computational performance. All of these parameters are defined in m_global.F90. If the users edit these parameters, re-compilation (make clean; make) is necessary.

Parameters

UC
Unit conversion coefficient from the computation unit (a mixed unit system using km for distance, g/cm for mass density, and km/s for seismic wave velocity) to the SI unit. If the user changes the unit of input, this parameter should be changed. The default values are 1e-15 for the 3D code and 1e-12 for the 2D codes.
MP
Preicsion constant for the finite difference calculation. If MP=DP (this is the default), the key calculation of the finite difference will be performed in double precision, while the other calculations are in single precision. If one changes it to MP=SP, all calculation will be done in single precision. In this case, the memory used in the 3D simulation is 2/3 as much as in the MP=DP case, and compuational speed also is faster than the default. On the other hand, single-precision calculation may lead numerical instability around the earthquake source after long-term integration. The values of DP and SP may differ at different compiler, but usually they take 8 and 4.
NM
Number of viscoelastic bodies adopted in the generalized Zener body. If this number is greater than 1, anelastic attenuation is automatically adjusted so that Q-values are approximately frequency independent in a frequency range specified by the paramreters fq_*. If this parameter is zero, the simulation is performed under the assumption of perfectly elastic body, and thus no intrinsic attenuation is taken into account. For 3D simulations, this parameter significantly affects the computation time and required memory size.