Simplifying access to High-Performance Galaxy Modelling

Welcoming page of the new GBKFIT web interface.


Current and upcoming large-scale spatially-resolved spectroscopic galaxy surveys such as SAMI, MaNGA, Hector, and WALLABY will revolutionise our understanding of galaxy formation by revealing the internal structure and kinematics of thousands of galaxies. While these unprecedented datasets are enabling entirely new avenues of investigation, they are introducing a demand for completely new analysis techniques which have yet to achieve wide-scale availability. Meanwhile, new and next-generation instruments like MUSE, GMTIFS, and HARMONI will — in the coming years — capture vast numbers of spectra during a single exposure, compounding the challenge of fully exploiting these data-rich products.


Using software support resources acquired under the 2017B semester of the ADACS Software Support scheme, Dr. Georgios Bekiaris of CSIRO Australia Telescope National Facility has partnered with ADACS to address these problems by creating a new web-based platform enabling quick-and-easy access to high-performance kinematic galaxy modelling tools for the international community.


ADACS has enabled an outward-facing user interface for GBKFIT: a publicly-available application for kinematic galaxy modelling. Features include a range of accepted data inputs (including flux, velocity, and velocity dispersion maps, as well as spectral cubes) and kinematic models (including exponential, flat, arctan, Epinat, and more), all wrapped in a secured web application (including a relational database and workflow management system) enabling providing support for a rich user community. The application provides GBKFIT users with an online platform for performing model fitting via a web browser. It consists of an easy-to-use web application providing tools for configuring and managing jobs; presenting results; and downloading results, while providing the computing resources needed to run this sophisticated modelling package.


“Many of the physical processes that take place in a galaxy are imprinted on the motions of its gas and stars. This makes galaxy kinematics an invaluable tool for studying galaxy structure, formation, and evolution.” says Dr. Bekiaris. “By modelling the internal motions of galaxies we can extract useful kinematic and morphological quantities which can improve our understanding of these wonderful celestial objects. Galaxy kinematic modelling can be a very computationally demanding process, both in terms of computational resources and run-time. The outcome of this ADACS project enables scientists to perform galaxy kinematic modelling analysis on remote supercomputing facilities, without requiring any technical expertise on parallel or distributed computing. Its intuitive kinematic analysis work flow now allows astronomers to increase their productivity by focusing on the science, while behind the scene, the use of supercomputers can accelerate the process of galaxy kinematic modelling, which in turn can accelerate the pace of scientific discovery.”

Snapshot of the workflow to create a kinematic analysis job
Portion of the result page showing GBKFIT output for an analysis job
All development has been released under the MIT license. The code is available at https://github.com/ADACS-Australia/ADACS-GBKFIT. After a request from the GBKFIT team, OzSTAR decided to host the platform. It will be made available publically soon. For more details, please contact Dr. Dany Vohl at dvohl@swin.edu.au.
This work was performed on the swinSTAR supercomputer at Swinburne University of Technology. All development was conducted by Dr. Dany Vohl, Dr. Shibli Saleheen, and Lewis Lakerink.