SKA-PST-DSP-TOOLS

This project provides Docker images for offline analysis of data recorded by the SKA Pulsar Timing (PST) backend. Two Docker images are built: ska-pst-psrchive and ska-pst-dspsr.

ska-pst-psrchive

The ska-pst-psrchive base image provides the PSRCHIVE suite of tools for the analysis of pulsar astronomical data in the form of phase-resolved averages known as folded pulse profiles. The installed tools include the development library, python bindings, and user-end executables. In addition to all of the build tools required to compile the code base, a number of third-party packages are also installed, including PGPLOT, FFTW3, CFITSIO, Armadillo, splinter, and yaml-cpp. Although not strictly dependencies, vim and gnuplot are also installed.

Pulsar-specific software packages (psrchive, tempo2, and psrcat) are installed in $HOME/Pulsar.

ska-pst-dspsr

This ska-pst-dspsr image builds on top of ska-pst-psrchive and provides the DSPSR suite of tools for the analysis of pulsar astronomical time series data. The installed tools include the development library, python bindings, and user-end executables. In addition to DSPSR, PSRDADA is also installed to help with managing PST voltage recorder data files.

Pulsar-specific software packages (dspsr and psrdada) are installed in $HOME/Pulsar.

ska-pst-dsp-demo

This ska-pst-dsp-demo image builds on top of ska-pst-dspsr, providing an example of how a Cmake based C++ application can use the libraries of DSPSR. This image requires that DSPSR in ska-pst-dspsr image is compiled with CUDA support (not currently the case). Once this has been updated, this image can be used as an example of how to deploy a CMake CUDA application.

ska_pst_dsp_demo

This is the example C++ application built by the ska-pst-dsp-demo image. This code is not built by default in the CI environment, but rather provided here as reference source code that derisked the integration steps for Cmake C++ code and auto-tools compiled DSPSR/PSRCHIVE/PSRDADA.