Dr Stef Salvini, Senior Researcher

Dr Stef Salvini                             

 

Dr Stef Salvini, Senior Researcher, talks to us about his work on the Square Kilometre Array, Moore's Law - and Monty Python!

 

When did you start at the Centre and what was your first role here?

I started in October 2006 at OSC (now ARC, Advanced Research Computing ) as Scientific Software Advisor.

Then, Anne Trefethen (then Centre Director) and I started a collaboration with Oxford Physics (Radio Astronomy). We were soon joined by Ben Mort and Fred Dulwich , and I became a Senior Researcher. The collaboration has since matured into our full involvement in the Square Kilometre Array (SKA) Project.

What is your background?

After an Italian Laurea (equivalent to BSc + MSc) in Chemistry, I obtained a PhD in Applied Mathematics and Theoretical Physics.

Then, I worked in industry in advanced High Performance Computing (HPC) for many years, developing new technologies and products as well as collaborating with international research teams and institutions in numerical and scientific parallel computing.


Summarise the research you are doing / your research interests in a few sentences.

I am currently interested in computational algorithms and their applications for Radio Astronomy within the SKA Project. A new generations of algorithms, more efficient than existing ones, is required by the next generation of Radio Telescopes. Some of the algorithms I have developed are now generally used by Radio Astronomers.

Why is this important to the scientific community and the world at large? 

The SKA is a truly international project, which will lead to the construction of two radio telescopes of unprecedented size and sensitivity: one in South Africa (medium radio frequencies) and one in Australia (low frequencies).

First, we are all very proud that such a very large international scientific project is centred in Africa.

The new instruments will enable scientists, amongst many other aims, to:

  • Investigate why galaxies exist . This requires investigations at very great distances, to study occurrences when the Universe was young and became transparent to radiation for the first time.

  • Test General Relativity, study pulsars, etc . General Relativity is currently a very accurate description of gravity.

  • Study the agglomeration of dust into increasingly larger sizes, from pebbles all the way to planets . Currently, the mechanisms for that are not clear.

  • Of course, there will also be unexpected findings – there always are and always will be when unprecedented experimental capabilities are created.

Pulsar

What would you like to do next, funding permitting?

See SKA to completion and able to deliver new science!

Are you involved in any wider collaborations? Why are these important?

SKA, of course! Only an international collaboration could achieve its ambitious targets. International collaborations give necessary diversity in outlook, opinions, experience: this is the lifeblood of a major scientific enterprise.

What publication /paper are you most proud of and why?

The invention of a new algorithm for the calibration of Radio Astronomical data. This decreased computing times by two orders of magnitude for existing instruments (from hours, to less than a minute), and more for SKA-class instruments. Radio Astronomers used to submit their computations overnight; now, it takes less time than making a cup of coffee! The algorithm is used widely, and will be used in SKA. I published the work with Stefan Wijnholds (ASTRON), which showed the validity of the algorithm: it is now known, jokingly, as "Stefcal" from our names ( A&A, 571, A97 ).

What do you think the most important issue s/challenges in your field will be in the next decade and how is the Centre placed to address them?

As Moore's Law comes to an end, carrying out increasingly complex, and costly, computations will require new algorithms, technologies and ways of thinking. The same will apply to other fields, of course. Cross-fertilisation between different areas of research will become more widespread, with ideas and approaches flowing much better than now.

 

Pythonscape What do you think the Centre does best?

The Centre provides a fertile ground for interdisciplinary thinking! Increasingly complex research involves a wide spectrum of skills; traditional departments are not always best suited to rise to the challenge. The Centre is and will increasingly be so.

Besides, it is fun to talk with and hear from people who work in different areas (I do love the Monty-Pythonesque "And now for something completely different" ...).