Solver Parameter Input XML File

The svFSI solver reads simulation parameters from a custom text format file. Simulation parameters are set using keyword/value pairs separated by a colon. A keyword consists of series of alphanumeric characters separated by spaces.

Keyword/value example

Coupled: 1
Min iterations: 1
Tolerance: 1e-6

Braces **{}** are used to define simulation parameters with sub-elements. For example the Domain keyword is associated with several other sub-elements of keyword/values pairs.

Domain: 0 {
  Equation: fluid
  Density: 1.0
  Viscosity: Constant {Value: 0.04}
  Backflow stabilization coefficient: 0.2
}

svFSIplus solver simulation parameters are stored in an XML-format file. The XML file organization and keyword names replicate the original text format using the following conversion rules

Spaces in keywords names are replaced by underscores
An additional value after a colon is replaced by an XML attribute used to identify the value. For example: Add equation: FSI is replaced by <Add_equation type="FSI" >.
Parameters with sub-elements are replaced by an XML element with sub-elements. For example Add equation: FSI { sub-elements } is replaced by <Add_equation type="FSI" > xml-sub-elements </Add_equation>.

Example: The original text

Add equation: FSI {
   Coupled: 1
   Min iterations: 1
   Max iterations: 10
   Tolerance: 1e-6
 
   Domain: 0 {
      Equation: fluid
      Density: 1.0
      Viscosity: Constant {Value: 0.04}
      Backflow stabilization coefficient: 0.2
   }
}

is replaced by new XML format with

<Add_equation type="FSI" >
   <Coupled> true </Coupled>
   <Min_iterations> 1 </Min_iterations>
   <Max_iterations> 10 </Max_iterations>
   <Tolerance> 1e-6 </Tolerance>
 
   <Domain id="0" >
      <Equation> fluid </Equation>
      <Density> 1.0 </Density>
      <Viscosity model="Constant" >
         <Value> 0.04 </Value>
      </Viscosity>
      <Backflow_stabilization_coefficient> 0.2 </Backflow_stabilization_coefficient>
   </Domain>
</Add_equation>

Pre-built svFSIplus Binaries

Pre-built svFSIplus binaries can be downloaded from the SimVascular SimTK downloads page. Installers are provided for MacOS and Ubuntu operating systems.

The compilers and other software used to build the binaries (listed on the downloads page) must be installed on the target platform in order to use the svFSIplus binary.

Building svFSIplus from Source

The svFSIplus source code can be downloaded and built on computers running MacOS and other Linux operating systems. Builds are most commonly done on MacOS and Ubuntu platforms.

The Windows operating system is currently not supported.

Build Process

Several software packages must be installed in order to build and use svFSIplus.

svFSIplus is built on a MacOS or Linux computer from a terminal using the following commands

mkdir svFSIplus-package
cd svFSIplus-package
git clone https://github.com/SimVascular/svFSIplus
mkdir build
cd build/
cmake ../svFSIplus/
make -j4

This creates the svFSIplus binary svFSIplus-package/build/svFSI-build/bin/svFSIplus*.

The binary can be installed in /usr/local using the following commands

cd svFSI-build
sudo make install

The svFSIplus* binary is now located in /usr/local/SV/bin.

It is convenient to update the PATH environment variable with the location of the svFSIplus* binary. Another option is to create an alias for svFSIplus. For example: alias svFsiPlus=/usr/local/SV/bin/svFSIplus

Required Software Packages

The following software packages are required to build svFSIplus

CMake
C++17 compiler
Visualization Toolkit
Open MPI

Software packages can be installed using package management tools like homebrew and apt.

CMake

svFSIplus is built using CMake. CMake is used to control the software compilation process using the CMakeLists.txt files located within the svFSIplus source. CMake first attempts to discover the software dependencies (e.g. C++ compiler) required to build svFSIplus. It will then generate the native makefiles used to compile it.

Visualization Toolkit

svFSIplus uses the The Visualization Toolkit (VTK) to read finite element mesh data stored in VTP and VTU formats. The VTP format is used to store 2D finite element meshes for inlet and outlet surfaces. The VTU format is used to store finite element volume meshes. svFSIplus can also write simulation results to VTK format files.

Open MPI

svFSIplus uses the Open MPI Message Passing Interface (MPI) libraries for parallel computing.

Using Trilinos with svFSIplus

By default svFSIplus uses custom linear algebra solvers (e.g. GMRES) in a simulation. It also supports several simple preconditioners.

svFSIplus can also be built with the linear algebra solvers and preconditioners provided by the Trilinos scientific software package. Using Trilinos can often reduce the execution time of a simulation.

Building Trilinos

The Trilinos libraries need to be built from the source downloaded from here. See the Quick configure, build and install hints for Trilinos for the instructions about how to build Trilinos.

Build with the third-party libraries (TPLs)

Boost
BLAS
HDF5
HYPRE
LAPACK
MUMPS

and with The Trilinos packages

Amesos
AztecOO
Epetra
EpetraEXT
Ifpack
ML
MueLU
ROL
Sacado
Teuchos
Zoltan

Compiling with Trilinos Libraries

svFSIplus is compiled with the Trilinos libraries using the CMake SV_USE_TRILINOS option

cmake ../svFSIplus/ -DSV_USE_TRILINOS:BOOL=ON

CMake will be able to find the Trilinos libraries once they have been built and installed.

Running a Simulation

The svFSIplus binary is run from the command line with an input solver parameters XML file as an argument

svFSIplus solver_params.xml

A simulation can be run in parallel on four processors using

mpiexec -np 4 svFSIplus solver_params.xml

Table of Contents