GOVERNMENT

The Higgs boson is elusive prey. Physicists have been searching for it since it was first proposed in the 1960s as a result of a theory to explain why all subatomic particles—and thus all objects in the universe—have mass. "The understanding of particle physics has reached a very advanced level, but there are still things that are puzzling," says Duke University physicist Ashutosh Kotwal. "We understand why particles and forces behave the way they do, and how they interact with each other. But our theories all only make sense if these fundamental particles have no mass, which is obviously not true." One way to give mass to the universe is to incorporate something new, called the Higgs field, into the equations. If the Higgs field exists, so do Higgs bosons, which physicists may be able to create in particle accelerators and measure with particle detectors. The scientists of the Collider Detector at Fermilab collaboration use the CDF detector and the Tevatron particle accelerator to search for and measure the Higgs and other fundamental particles and forces. But particle physics discoveries and measurements don't just require machinery and ingenuity; today it also helps to have the computing power of the grid. CDF physicists use the grid for an essential tool in the hunt for the Higgs—Monte Carlo simulations. "Simulation jobs are well-suited for being run on the grid," explains Pavel Murat, a physicist from the Fermi National Accelerator Laboratory. "They don't require a lot of data movement, but are CPU intensive." Monte Carlo simulation of the simultaneous production of a Higgs boson and a Z bozon, one of the predicted signatures of the Higgs boson. The Z boson decays to two electrons and the Higgs boson decays to two b quark jets, all with large transverse momenta. Image Courtesy Ashutosh Kotwal For the Higgs boson, theoretical physicists predict how the particle might be created, and how it might decay, and experimental physicists model what those events would look like when viewed through the CDF detector. Those theoretical and detector models are then used to simulate millions of particle collisions in the CDF detector. Collected data from the detector is then compared with simulated data to see if any real events match the predicted Higgs events. The millions of simulated events are also used by physicists to learn how to separate out the interesting real events—like Higgs bosons—from the background, or detector signals coming from well-known phenomena. For two years, CDF has been using a network of computers worldwide to simulate the hundreds of millions of events per year that are required to make discoveries and measurements. And for the last six months, they've been using the grid to help simplify job submission and management. Through the Open Science Grid, simulations are being run on three CDF sites in North America, with three more—including one EGEE site—to be added within the next few months. "The tools to direct jobs to different sites come from the OSG," explains Kotwal. "We developed the user interface, a thin layer on top of the OSG software." CDF simulations on the grid will also be extended to European sites. A group of collaborators in Italy is extending the grid simulations to sites in Europe that use the LHC Computing Grid software stack. That project would keep the same CDF user interface but use LCG tools behind the scenes. So will CDF discover the Higgs boson? By combining their collaboration's findings with that of the DZero collaboration—the other large experiment at the Tevatron—there is a chance that first experimental evidence for the Higgs boson could be announced within the next few years. And another hunt is on the horizon: physicists at the Large Hadron Collider, the next-generation particle collider currently being built at the CERN laboratory in Geneva, Switzerland, will begin their own search for the elusive particle at the end of 2007. For more information visit the CDF Web site and Open Science Grid Web sites. —Katie Yurkewicz, Science Grid This Week