CFI Innovation Fund series: researching the fundamental structure of nature

Simon Fraser University physicist Michel Vetterli is leading an SFU research project that has received $3.5 million from the Canada Foundation for Innovation (CFI). The ATLAS Tier-1 Data Centre project is one of five SFU-led research projects benefiting from the CFI Innovation Fund.

The ATLAS Tier-1 Data Centre provides large-scale, coordinated resources to the Worldwide LHC Computing Grid (WLCG) to enable ATLAS data analysis. The recent CFI award will support the high-performance computing needed to analyze the enormous amount of data generated by ATLAS, while developments in globally distributed computing capabilities filter down to everyday use by society.

Vetterli, a big data expert, is participating in the Data Visionaries lecture series. Presented by KEY, SFU’s Big Data initiative, this free, public event is on Wednesday, Nov. 22. Vetterli’s talk is entitled ATLAS Computing: Managing and Analyzing Petabytes of Data on the Worldwide LHC Computing Grid.

SFU News spoke with Vetterli about the project and the international collaboration between ATLAS researchers who are studying extremely high-energy particle collisions. Their goal: to learn about the fundamental building blocks of matter and the forces of nature.

What would you most like the public to know about your research project and goals?

First, on the physics side, the search for an understanding of how the universe works at the most fundamental level is one of the grandest challenges facing humanity. The potential spin-offs of this type of research are not immediate, but history has shown that today’s abstract ideas lead to transformational applications of the future. Tomorrow’s innovations are based on today’s basic science.

Second, concerning the specific research supported by the CFI Innovation Fund, work done at the cutting-edge of globally distributed computing has led—and will continue to lead—to better tools for society’s general use. Canada is a significant player in both these areas of research and CFI funding has been crucial to our participation as an equal partner on the world stage.

Tell us about your academic background, research interests and what led you to where you are today in your career.

I received a BSc from McGill in honours physics and went to McMaster University for postgraduate studies, leading to a PhD in nuclear physics in 1985. I then went to TRIUMF —Canada’s National Laboratory for Particle and Nuclear Physics—for a postdoctoral position and was hired as a lab scientist in 1989.

While at TRIUMF, I worked in intermediate energy nuclear physics and used Vancouver as a base to participate in the HERMES experiment at DESY, an international research centre in Hamburg, Germany. I studied the structure of the protons and neutrons that make up the atomic nucleus.  

I moved on to SFU in 2001 and, at the same time, joined the ATLAS collaboration at CERN in Geneva, Switzerland. ATLAS is a multi-purpose detector for studying proton-proton collisions at the highest energy ever produced in the laboratory. The Run-1 physics program (2010-2012) culminated in the discovery of the Higgs boson in 2012, which led to the 2013 Nobel Prize in physics. Run-2 (2015-2018) is allowing us to study the properties of the Higgs boson with high precision, as well as search for new physics beyond the standard model of particle physics.

In parallel to these physics activities, I have been heavily involved in large-scale computing. I am a co-founder of WestGrid, a network of high-performance computing centres—one of which is at SFU—to support academic research in western Canada. This led to the establishment of the national Compute Canada infrastructure. 

What are some of your key research accomplishments and how are they informing your future research?

The HERMES experiment at DESY in Hamburg made ground-breaking measurements of the structure of protons and neutrons—the constituents of the atomic nucleus. In particular, we made the best determination at that time of how the property called the spin of the proton is related to the spin of its constituents. These are textbook results.

The more recent discovery of the Higgs boson at ATLAS—and its sister experiment (CMS) at the Large Hadron Collider—gave definitive evidence of the existence of the Higgs field and therefore confirmed our understanding of how the mass of subatomic particles is generated.

Without what we now know as the Brout-Endler-Higgs mechanism, no bound states of subatomic particles would exist, and the universe would be a very different place without atoms or nuclei.  

I am also proud of the work I have done in high-performance computing, both in Canada (WestGrid, ATLAS Tier-1 Data Centre) and internationally (Worldwide LHC Computing Grid).

Please explain the fundamental value of your research and why it is important to Canadians and global citizens.

The goal of particle physics is to identify the basic building blocks of matter and how they interact. Beyond its intrinsic importance, a clear understanding of these fundamental problems inevitably leads to transformative spin-offs.  

There are numerous historical examples, including cathode-ray tubes (television), nuclear medicine, and even the world-wide web that was developed at CERN to permit worldwide data-sharing and collaboration.

By understanding the world at its most basic level, we can better control it for society’s benefit. Even in the absence of technical spin-offs, Canadians owe it to themselves—and to the global community—to participate in this quest for knowledge of how our world really works.

Can you tell us about your proposal’s research agenda?

The CFI award will support the upgrade and expansion of the ATLAS-Canada Tier-1 Data Centre in Vancouver. The ATLAS experiment generates an enormous amount of data because it searches for extremely rare phenomena in the collisions of high-energy protons. The primary datasets exceed seven petabytes of data (seven million GB) per year and are augmented by a factor of 2-3 by secondary datasets.

All of these data must be reprocessed and analyzed multiple times, which presents a very significant computing challenge. The solution chosen is to build an international network of high-performance computing centres that are connected by high-speed networks, and process the data in an organized manner, essentially acting as one coordinated system.  

This Worldwide LHC Computing Grid (WLCG) is organized into multiple tiers that fulfill overlapping but different functions in a staged approach to the analysis. In particular, the Tier-1 layer stores a copy of the raw data from the experiments and reprocesses them when new information is available or better algorithms are developed.  

One of the 10 Tier-1 centres worldwide is in Vancouver, now at TRIUMF, but in the process of moving to SFU. The recent CFI grant is to replace aging equipment to ensure the system’s reliability, as well as to expand the centre to handle the ongoing data collection at CERN.

These new resources will help us analyze and mine the new data—as well as the old—to refine the searches for new physics phenomena beyond the Standard Model of particle physics.

How is your research innovative and cutting-edge, and why is it suitable for CFI funding?

Large-scale distributed computing has been used for years to attack problems such as, for example, the Search for Extraterrestrial Intelligence (SETI). The SETI-at-home program uses desktop computers in people’s homes and businesses to analyze data collected by radio telescopes. Analyzing ATLAS data requires a much more coordinated effort than the SETI project, so that the tools developed are more sophisticated.  

Of particular note is the enormous dataset that must be managed (hundreds of petabytes), which is unprecedented in large-scale distributed computing.

The tools developed by the WLCG to solve this challenge can be used to help attack the big data problems that are becoming more and more prevalent in society. The real innovation of the WLCG is its character as a data grid, not just a computational grid. What is learned in the grid environment can help develop the computing cloud that can be accessed by single users.

This type of tool development and deployment is perfect for public-sector funding because the benefits are important for improving the tools used by society at large.