PaCER Seminar: Radio astronomy
10:00am - 11:00am
Ten research projects were successfully granted access to the first Pawsey Centre for Extreme-scale Readiness (PaCER) program, establishing Australia’s research platform for extreme-scale computing.
Pawsey is hosting a series of seminars throughout June showcasing the first cohort of PaCER researchers’ projects. The last event in the series focuses on Radio Astronomy and showcases Melanie Johnston-Hollitt from Curtin University and Martin Meyer and Marcin Sokolowski from the ICRAR UWA and Curtin node respectively.
The focus of the PaCER program is on both extreme scale research (algorithms design, code optimisation, application and workflow readiness) and using the computational infrastructure to facilitate research for producing world-class scientific outcomes.
The program is a partnership for collaboration between researchers and Pawsey Supercomputing Centre supercomputing specialists using the latest infrastructure provided by Setonix.
Join us each Tuesday in June to find out about their projects and their research impacts across Australia and the world.
About the projects
- Parallel Interferometric GPU Imaging
- Melanie Johnston-Hollitt – Curtin University
Collaborators: Curtin University/ University of Toronto/ Pawsey Supercomputing Centre
Johnston-Hollitt’s project, Parallel interferometric GPU imaging, aims to combine the distributed nature of recent interferometric reconstruction algorithms with fast instrumental modelling using GPUs to accurately reconstruct images from extremely large data sets for future instruments such as SKA_Low
To provide the first end-to-end exascale astronomy calibration and imaging pipeline which has been optimized for HPC, the team will scale up calibration pipelines and deploy them in a multi-node GPU scenario combined with modern imaging algorithms scaled and deployed on HPC.
- HiVIS – HI Visibility Imaging for the SKA
- Delivery of a next-generation data storage approach to unlock deep SKA and pathfinder observations
- Martin Meyer – ICRAR UWA
Collaborators: ICRAR / CSIRO Astronomy and Space Science / SKA Organisation Headquarters (SKAO)/ Oak Ridge National Laboratory
Meyer’s project addresses one of the most significant outstanding Grand Challenge Problems for the SKA Observatory – how to optimally image multi-day deep datasets.
By developing a sparse data storage and processing pipeline based on UV-grids, this project aims to reduce the visibility storage requirements for these projects by an order of magnitude. The developed methods will simultaneously enable critically needed reprocessing to optimise the scientific outcomes from these datasets and opens up the possibility for higher resolution spatial and spectral imaging than would otherwise be possible
As a testbed, Pawsey will be used to image 250h of ASKAP (Australia Square Kilometer Array Pathfinder) data from the DINGO pilot surveys, along with 500h of its first ultradeep field. These will yield some of the deepest images ever taken of atomic hydrogen content in the Universe, enabling ground-breaking novel studies of the role this fundamental fuel has played in the ongoing evolution of galaxies and its connection to their dark matter halos.
In addition, the results will demonstrate a solution for the SKA data challenges in deep imaging.
- BLINK (Breakthrough Low-latency Imaging with Next-generation Kernels)
- BLINK and you’ll miss it: blazingly-fast all-sky radio astronomy pipelines
- Marcin Sokolowski – ICRAR Curtin University
This project aims to bring Pawsey’s latest-generation hardware to make real-time, image-based transient searches feasible, applied to the Murchison Widefield Array, the foremost precursor telescope for the low-frequency component of the Square Kilometer Array.
Making real-time, image-based transient searches feasible will require a combination of the latest data processing technology offered by modern supercomputers, and novel data processing algorithms that have been optimised for both speed and sensitivity to transient signals.
If the computational challenges in this area are overcome, the untapped potential to form widefield images with millisecond time and kilohertz frequency resolutions coming from instruments such as the recently-upgraded Murchison Widefield Array (MWA) and the upcoming low-frequency Square Kilometre Array (SKA-Low) will transform these telescopes into extremely powerful instruments capable of detecting Fast Radio Bursts (FRBs), other fast transients like “rotating radio transients” (RRATs), i.e. sporadically emitting neutron stars and possibly other phenomena in real-time.
About the researchers
Melanie Johnston-Hollitt is an internationally prominent radio astronomer working in the space between astrophysics, computer science, and big data. She is Director of the Curtin Institute for Computation (CIC), a knowledge accelerator based on the use of data science and high-performance computing aimed at producing better outcomes for research, government, and industry. In her 20-year career, Melanie has been involved in the design, construction, operation, and governance of several major international radio telescopes including the Low Frequency Array in the Netherlands, the Murchison Widefield Array (MWA) in Western Australia, and the billion-dollar Square Kilometre Array (SKA) which will be hosted in both Australia and South Africa. She was a founding member of the Board of Directors for the SKA Organisation where she worked on the international governance, scientific, and technical aspects of the telescope. In particular, she led the Science Analysis Pipeline design and contributed to the Science Data Processor pipeline which will be needed to extract knowledge from the vast amounts of data the SKA telescope will generate. She is the immediate past director of the 65 million-dollar MWA radio telescope and spent 9 years involved in the project, including 6 years on the international executive board (4 years as board chair) and 3 as director. As MWA Director she oversaw the third phase of the MWA project realised via the design and funding for the new ‘MWAX correlator’ – a GPU-based bespoke compute system. Her research interests span the intersection between radio astronomy, signal and image processing, and big data analytics and she leads both the multi-disciplinary team of data scientists in the CIC and the galactic and extragalactic science team in the Curtin Institute for Radio Astronomy who are exploiting the MWA and other telescopes to uncover the mysteries of the Universe.
Martin Meyer’s research focuses on surveys for neutral atomic hydrogen (HI) and the role played by hydrogen gas in the formation and evolution of galaxies. Matin leads the DINGO survey, a project that will take deep HI observations with the Australian SKA Pathfinder to understand how the HI content of the Universe has evolved over the past 4 billion years. In the lead-up to this project, He is working on CHILES, a deep HI survey being carried out with the VLA, as well as wide-field HI stacking experiments in the GAMA G09 field also being observed with this facility. Martin is a member of the SKA HI galaxy science working group.
Marcin Sokolowski is a Research Fellow at Curtin Institute of Radio Astronomy (CIRA), Australia.
Marcin received Master Degree in Physics from the University of Warsaw, Poland in 1998and a Ph.D. degree in Physics from the National Centre for Nuclear Research (NCBJ) in Warsaw, Poland in 2008. Between 1998 and 2003 he worked as a software programmer in a software company. From 2008 until 2012 he was an Assistant Professor in NCBJ where we worked on software for the automation of wide-field robotic telescopes and algorithms for the identification of optical transients of astrophysical origin.
In the last few years, he has been looking for Fast Radio Bursts (FRBs) and other transients at low radio-frequencies using the wide-field images from the Murchison Widefield Array (MWA) and all-sky images from prototype stations of the low-frequency Square Kilometre Array (SKA-Low).
He works in a large international team of engineers and scientists on verification, commissioning and developing novel calibration methods for prototype stations of the SKA-Low to be built in Western Australia in the coming years.