Feature - People behind the LHC Grid: A Bird's eye view
Ian Bird is project leader for the Large Hadron Collider Computing Grid, where he is responsible for grid deployment and all of the technical work involved in setting up the grid infrastructure for the LHC.
iSGTW: If you were at a cocktail party, and someone asked "What do you do," how would you answer, in lay person's terms?
IB: Well, I'd try to change the subject.
Seriously, I try to coordinate the overall activities of the LCG-the experiments, the service providers, the money and the people. I report to CERN management and oversight, to make sure all the nuts and bolts are in the right place. And, of course, I think about all the technical aspects.
iSGTW: How do things look now?
IB: Well, we're very upbeat. Of course, we always knew that there would be surprises, but we are handling them. As has been the case for the past several years the grid is in constant use for simulations and is used for evaluating the data coming in all the time from the naturally occurring cosmic rays that are hitting the detectors-and this is helping to calibrate the equipment.
iSGTW: Did you always have an interest in physics and computing?
IB: Believe it or not, but when I was in school, my headmaster was not very encouraging about my getting into physics. He told me to get into something else.
But that didn't deter me-I always wanted to know how the world works. That's the beauty of physics, that it goes into the how and the why.
So, I pursued physics-and computer science-anyway. And that's the message I try to pass on to my children about their choice of careers: do what you are interested in.
As for computing, ever since I started as a physicist, I was always involved in the software for the experiments I was working on. In the beginning of the '90s, we had one of the very first SUN clusters, and we were one of the earliest to use CONDOR .
iSGTW: How long have you been at CERN?
IB: I first came to CERN in 1977, as a graduate student, to work in the BEBC Bubble Chamber. Then I was back in 1981 as a post-doc with the European Muon Collaboration experiment. I then did a series of collaborations into the 1990s, on projects such as NOMAD, and then I went to Jefferson Lab in the States in 1996. In 2002 I came back to CERN, and then took over from Les Robertson after he stepped down earlier this year.
And Les was really one of the key people behind LHC computing and the grid. He got the funding and the instrumentation, and got all the governing bodies on board.
In the beginning
iSGTW: What was it like in the early days?
IB: In the year 2000, we had a model for handling the vast amount of data we knew the LHC would generate, which already featured the structure of tiers and high speed networks. At that time the new grid technology suggested a way to implement the model, while advances in network technology solved the throughput problem; we could just buy off-the-shelf products for the network.
iSGTW: What do you think the grid needs to do to become more widely accepted?
IB: The current grid middleware is not very user-friendly, it's a bit baroque not only for end-users but for site managers as well. That's probably the biggest problem. But any new technology has to go through this phase; the difference in our case being that we're trying to use it now in large-scale production. But over time, it should become easier to use and more robust.
iSGTW: In your mind, what do you think we will find?
IB: I can't even imagine it. Whatever it is, we're ready.
iSGTW: Any parting thoughts?
IB: This scale of collaboration has never really been done before in computing. The technology is one thing, but getting all these centers to work together is quite an achievement.
What CERN has going for it is that it has always been about international collaboration. The science transcends the politics.
-Dan Drollette, iSGTW