Life is hard, and operating systems are even worse. The current prerequisites for SFXC on Debian 12 are given in the [Apptainer recipe] in the source repository as:
apt-get install -y python3 python3-pip build-essential git flex bison \
openmpi-bin libopenmpi-dev python-is-python3 \
libfftw3-dev libgsl-dev gfortran-multilib g++-multilib
As there are some doubts whether the CALC10 software used by SFXC is 64-bit safe, the model generation tools are compiled in 32-bit mode. This means that if you are using a 64-bit (x86_64) Linux system, you will have to install 32-bit versions of the C++ and Fortran compilers and their support libraries; these packages are included in the above list.
The Apptainer recipe is actively maintained and should be considered definitive. Finding equivalent packages for other flavours of Linux and Unix (including Apple systems) is left as an exercise for the interested reader.
The SFXC software correlator can be distributed under the terms of the General Public License (GPL). The SFXC Git repository can be found at [[https://code.jive.eu/JIVE/sfxc]]
The repository can be downloaded using:
git clone https://code.jive.eu/JIVE/sfxc.git
The current production release is taken from the stable-5.1 branch which can be checked out using:
git checkout stable-5.1
In principle this branch will only receive bug fixes. Development of new features happens on the master branch, which can be checked out using:
git checkout master
Note that using that version is not recommended unless you feel like being a guinea pig for testing the bugs we introduce while adding all the new cool stuff.
Building SFXC should be as easy as:
cd sfxc
./compile.sh
./configure CXX=mpicxx
make
make install
This will install SFXC in /usr/local, which typically requires
super-user privileges and may not be what you want. You can specify a
different location by using the --prefix option when running
configure.
SFXC can optionally use the Intel Performance Primitives (IPP) library.
See the --enable-ipp and --with-ipp-path options to configure.
SFXC comes with a couple of GUI tools to visualize the correlation
results. This needs some additional Python packages, known to apt as
python3-ply and python3-qwt5, along with the Python VEX parser
module that can be found in the vex/ top-level subdirectory. This
module uses a standard Python distutils setup.py, so can be installed
by:
cd vex
python setup.py build
python setup.py install
The last command will probably require root priviliges; setup.py
offers a couple of alternative installation methods that avoid this.
More information on the VEX parser is provided in vex/README.
To convert the SFXC correlator output into FITS-IDI, additional tools are needed. Information on how to obtain and build these tools is available at https://code.jive.eu/verkout/jive-casa.
A very basic User’s Manual for SFXC can be found here.
We kindly request users of SFXC to reference our paper that describes its algorithm and implementation The SFXC software correlator for very long baseline interferometry: algorithms and implementation, A. Keimpema et. al., Experimental Astronomy, Volume 39, Issue 2, pp.259-279.
Content built by Des Small (JIVE).
Built with ♥ — Markdown + HTML + CSS + Prism.js + a bit of AI + Jack Radcliffe (2025)