From animation movie to particle accelerator

From animation movie to particle accelerator

Opening Academic Year 2022: Hustinx Prize for computer scientist Daniel Cámpora

05-09-2022 · Interview

MAASTRICHT. What started as an idea when he was a student, grew to be an important part of one of the world’s largest and most expensive experiments. During the opening of the academic year, Daniel Cámpora received the Edmond Hustinx Prize of fifteen thousand euro for his work on the interface between computer science and particle physics.

“I like to make things faster,” Spanish Daniel Cámpora summarises his work. Those things are computers. Or to be more precise: computers that have to independently analyse and process large amounts of data.

In the past years, this has resulted in a “record-breaking system in the world of particle physics,” say the Maastricht scientists who nominated him for the Hustinx Prize. They referred to the so-called Allen programme, that Cámpora came up with and developed together with his colleagues. At the moment, it forms an important part of a large experiment with the CERN particle accelerator in Geneva, the institute where Cámpora worked for ten years, before he came to Maastricht two years ago – after a brief interlude at Nikhef in Amsterdam.

While he is not even a physicist, he emphasises. During his training to become a computer scientist in Sevilla, he didn’t take one single course about particle physics. So, how did he end up at the renowned CERN institute? “The particle accelerator produces a tremendous amount of data, which is all processed by computers. I thought it would be interesting to work with them, so I applied for an internship.” Which he got, followed by another and eventually a PhD position. “Because of my background, I saw possibilities that physicists didn’t think of.”

Convincing

One of these was completing calculations on graphic cards (so-called GPU’s), a job that at CERN is usually carried out by processors (CPU’s). “When I started at CERN, graphic cards were already being used to an increasing extent, for example to create animation movies and computer games. Unlike processors, they can complete multiple calculations simultaneously. I realised that this also enabled them to process the data from the particle accelerator much faster.”

His supervisor gave him the space to work on his idea. “It was great that I was given that opportunity. It took a lot of time and energy, and success was not guaranteed.” Managing a team of mainly students and PhD candidates, he developed the necessary software, which eventually resulted in ‘Allen’. After about two years of tinkering, the programme appeared to be so successful that other scientists became interested. “But convincing everyone to actually use your new system in experiments, is a completely different story. The particle accelerator is a very expensive machine, and a lot of research depends on it. Everything has to work without fail from the first moment.”

Exciting

Eventually, Cámpora and his colleagues succeeded in eliminating all doubt. “The system turned out to work two to three times more efficiently than the ‘traditional’ systems. In other words, it can deal with a larger influx of data. It also uses up to about seventy per cent less energy and it is cheaper.” This appeared to be so convincing that Allen became responsible for the data processing of one of the four large experiments of the Large Hadron Collider (LHC), the world’s largest and most powerful particle accelerator (see box).

This started up again last April, after more than three inactive years when the machine received an update. An exciting period for Cámpora. “We are the first ones to use graphic cards on this scale for data filtering. I do feel pressure in that respect, but in a positive way. When I wake up in the morning, I can’t wait to see if everything is still running well.” So far, that has been the case. “It can already be called quite groundbreaking. The CERN community has embraced the programme. It is also attracting a lot of talent, who want to see how it works.”

He wants to use the prize money to improve the software even further. “The LHC is constantly in development, with regard to technology it continues to be state-of-the-art. Allen should work even faster and better in the future.” He also sees possibilities outside CERN. “For example, large observatories, or systems that predict the weather. These also deal with large amounts of data, and our programme can accelerate processes.”

Looking more efficiently for antimatter

The Large Hadron Collider particle accelerator makes elementary particles collide with each other at gigantic speed. With every collision particles are produced, which can put physicists on the trail of unknown physics. For example, they are looking for so-called beauty quarks, which might unveil why the universe is made up of matter, while there is no trace of its counterpart: antimatter.

As the detectors register a gigantic number of particles whizzing by, it is impractical to record all these measurements, which would fill about eight hard drives per second. “Fortunately, most particles behave in a way that we already understand,” says Cámpora. “The computer immediately filters this ‘uninteresting’ information out. This requires a tremendous amount of computing power. The software has to reconstruct all the particle collisions from a tangle of data. After which it independently determines which data could indicate new physics. It is of the greatest importance that a programme such as Allen functions properly: everything that it doesn’t single out is lost forever.”

This appears to be successful: not only does Allen work faster, but it filters the information more precisely too. “This enables physicists to carry out research more efficiently.”

Edmond Hustinx Prize

The Edmond Hustinx Prize for research is intended to “underscore the meaning of science for society” and to accentuate the importance of Maastricht University for Limburg. The prize – 15,000 euro – goes to a different faculty every year, which this time is the Faculty of Science and Engineering (FSE).

Photo: Joey Roberts

Tags: OAY2022,hustinxprize,particle,physics,computer science,fse,openingacademicyear,science,instagram

Add Response

Click here for our privacy statement.

Since January 2022, Observant only publishes comments of people whose name is known to the editors.