Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

It's about knowing whom to talk to

Share this on social media:

Topic tags: 

John Murphy profiles the Head of Numerical Relativity Group, Albert Einstein Institute

To most people, working out what happens when two black holes collide in space seems a strange thing to do for a living. What sounds like 'The Gods Playing Billiards' is an urgent and pressing task for Edward Seidel of the Max Planck Institute for Gravitational Physics, Albert Einstein Institute, in Potsdam, Germany. As experimental astronomers around the world prepare to look for gravitational waves, his team is using numerical solutions to Einstein's equations of General Relativity to suggest what they should be looking for. There are so many black holes out there that collisions probably happen a few times a year, if only we could detect them.

Seidel's work is not confined to theory. He is also developing the technology that will allow experimental and theoretical groups to share data and bring to bear on it the computing power that will be required to interpret that data. Professor Bernard Schutz, director of the Albert Einstein Institute, said: 'He is unique in the field of numerical research into Einstein's equations, firstly because of his breadth, from understanding the mathematical theory, right though to implementing the theory and the computer science of applications. His other strong quality is his personality, which people gravitate to. He motivates a team of people very well to work together. He also has a phenomenal memory about who is doing what and who is a specialist in what. If you watch his e-mail, you see him constantly being asked by people who they should talk to about something and he always knows someone who they should get together with to talk about something.

'His group is probably the most advanced in terms of numerical solutions to the problem of collisions of two black holes, which is a key problem because it has no known parallel in Newtonian gravitational theory. We expect to be able to observe gravitational waves from these events, so the more we understand about them from simulation the easier it will be to recognise them. He has led the largest group in this area and the group that has made all the breakthroughs. 'His group has also achieved some great things in grid technology, particularly in the use of widely separated computers to work on the same problem simultaneously. They have worked on the key problems. Seidel does not worry about the small things that other people can clean up, he worries about the key problems.'

Seidel was born in Bethesda Naval Hospital in Maryland. His German colleagues often think that because of his name he is a native German. In fact his family moved to the US in the 19th Century and he has only lived in Germany for six years. His father was a Captain in the US Navy. By the time Seidel came on the scene his father had gone back to college to get a PhD and was a psychology professor.

He grew up in awe of the NASA space programme and as a child longed to become an astronaut. When he discovered this would not happen because he needed to wear glasses, his interest turned to electronics and as a teenager he built his own synthesiser.

His early college years were marked by rebellion, having proved a sucker for all the temptations and distractions of college life. He was suspended in his first year for 'disciplinary reasons' and by the end of the second year was ranked 970th out of 1000 in his year. He decided it was time to take a break and headed to Switzerland, where he intended to work in tourism and spend time skiing. His job lasted just six weeks and he headed off to travel around Europe before ending up back in a dead end job in South Carolina. He decided that if he was ever to amount to anything he needed to get his act together and went back to college - this time to work.

In his last three years at college, he scored highly on the GPA scale and, after graduating, was able to get into a graduate programme first at the University of Pennsylvania and then at Yale after he had discovered theoretical astrophysics as his true calling.

He had already developed an interest in numerical methods but, at the time, there was a severe shortage of the really big iron needed to do this work. At around this point, Larry Smarr was forming the National Centre for Supercomputing Applications (NCSA) at the University of Illinois. Smarr had pioneered work in solving Einstein's equations of General Relativity for the collision of two black holes and it was the need for more horsepower that drove him to get academic access to the large supercomputers that were then controlled by the military. Along the way, he created a high-speed backbone network that evolved in to the Internet.

Smarr had formed a group to work on numerical solutions to the 'two black hole' problem and Seidel won a place on the team under David Hobill. Within a few years, Hobill had left and Seidel ended up in charge of the team. By 1996 Seidel had reached a tenured position at the University of Illinois, but his work with Smarr, Hobill and others had projected his reputation to an international audience and he was regarded as 'up and coming'.

At the same time, Bernard Schutz had persuaded the Max Planck Society that it should back a group doing research into General Relativity (if for no better reason than Germany was Einstein's birthplace). The Society agreed to fund the formation of the Albert Einstein Institute and, with development encouraged in the former East after the fall of the Berlin Wall, based it in Potsdam. Schutz had spotted Seidel and was able to offer him an unparalleled package in terms of money and facilities. He could form his own team and would have brand new lab facilities - whatever he could want. The offer was too good to turn down and Seidel moved his family to Berlin in 1996, together with some members of his team from NCSA.

He said: 'My young son was excited by the idea, but my slightly older daughter said she would commit suicide if we moved. But six years on it's my son who talks about moving back to the US and my daughter who wants to stay.'

Things were not totally rosy. Being based in former East Germany meant it was sometimes difficult for staff to get phone lines connected and, while the main Albert Einstein Institute has a new building, Seidel has found himself using a former Russian Army barracks for some of his work.

But the resources he was offered came through. Seidel has found a shortage of large-scale computing facilities in Europe similar to that in the US before the NCSA was formed. Part of his work today is to try to get international co-operation through the European Union. However, because of his contacts and through retaining his position at the University of Illinois, he has been able to get time on the large US facilities for his team's work. He is also still able to apply for grant support for projects from US funding bodies.

Having got used to working across long distances and making use of far-distant computing resources, Seidel's group decided not just to work on their scientific problems but also to contribute to the computer science effort in grid technology. Its most famous result is the Cactus Code Server, which simplifies the parallel use of different machines to run a large application. With teams working to detect gravity waves being based all over the globe, it was important to him that they would work together. Given the scale of gravity waves, it should be expected that anything detected on one side of the Earth should be simultaneously detected on the other, so experimentalists and theoreticians need to be in close contact to make sure that what they are seeing is not due to seismic activity or a passing truck. Seidel said he is building good relationships with the experimentalists.

He said: 'In fact we get on better with the experimental groups than they do with each other. There will be Nobel Prizes at stake in this area. I believe that gravity-wave astronomy will become as important in the next 50 years as radio astronomy has been in the past 50. Before radio astronomy we had no idea about quasars and so there should be some surprises when we start looking at gravity waves. What we are doing at the moment is trying to imagine what we will see, so that the experimenters know what they should be looking for, but we also need to see the live data so that we can start thinking about how to interpret it.'

Seidel's group attracted a lot of attention at the recent Supercomputing 2002 conference, when they demonstrated their Global Grid Testbed Collaboration, involving 70 machines in 15 countries on five continents, running four different applications where each machine on the grid was handling part of the same application. The machines ranged from a 1024 Linux cluster to a Sony Playstation!

Curriculum Vitae

1982
The College of William and Mary, Williamsburg, VA, BS in Mathematics and Physics with High Honors in Physics
1983
The University of Pennsylvania, MS in Physics
1988
Yale University, New Haven, CT, PhD in Relativistic Astrophysics
1987-1989
Postdoctoral Research Associate, Department of Physics and McDonnell Center for the Space Sciences, Washington University
1989-1991
Postdoctoral Research Associate, National Center for Supercomputing Applications (NCSA), University of Illinois
1993-96
Research Scientist, NCSA
1994-present
Senior Research Scientist, NCSA
1993-96
Visiting Associate Professor of Physics, University of Illinois
1996
Associate Professor of Physics and Astronomy with tenure, University of Illinois
1996-present
Adjunct Professor, Depts. of Astronomy and Physics, University of Illinois
1996-present
Head of Numerical Relativity Group, Albert Einstein Institute, Max-Planck-Institut für Gravitationsphysik, Potsdam