Feature: Worldwide Grids, Worldwide Science
Scientists at last week’s meeting of the American Association for the Advancement of Science in San Francisco discussed the use of grid technologies and volunteer computing to fight disease, predict earthquake effects and hazardous weather conditions, understand the origins of the universe, and decode our own behavior.
“Science is distributed because life is distributed,” said TeraGrid Director Charlie Catlett in one of three sessions devoted to grid computing. “The Internet is worldwide, people are putting computers and storage facilities on the Net. The question is not whether you want a distributed infrastructure, but whether you want to use that infrastructure to do science.”
During the grid-focused sessions attendees learned about the global growth of distributed computing infrastructures and their use by scientists around the world.
The international theme was continued by Tony Hey, of Microsoft, and William Chang from the National Science Foundation’s Beijing office. Hey presented the state of grid computing and e-Science in the UK and Europe, while Chang discussed the development of networking and grid computing in Asia, including Chinese grid initiatives and the PRAGMA project, which bridges the nations of the Pacific Rim.
The majority of the speakers at the three sessions presented ways that distributed infrastructures are being used today for science.
Particle physicist JoAnne Hewett from the Stanford Linear Accelerator Center discussed modern physics research and the use of grid infrastructures. Discussing potential discoveries in particle physics that might result from the new experiments at the Large Hadron Collider and the LHC Computing Grid, she noted that these advances are made possible only by leaps in accelerator in computing technologies that have happened in the past decades.
“In 1933, the 11-inch cyclotron, an early particle accelerator, was invented at Berkeley and the IBM Tabulator was used for computation,” said Hewett. “Today we have Terascale computers that are more than 100 billion times more powerful than the Tabulator, and the LHC is more than 100 billion times more powerful than the 11-inch cyclotron.”
Physical, biological and earth sciences dominated the grid applications on display at the AAAS sessions; however, the use of grids for social science was also featured. Bennett Bertenthal from Indiana University spoke about the Social Informatics Data Grid, an initiative to help social scientists collaborate and share data, often for the first time. SIDGrid data is quite different from that used by physicists or chemists–largely in the form of streaming, time-sequenced data such as video and audio.
“We see the SIDgrid as transforming our science,” said Bertenthal.
The use of distributed computing for education was also highlighted. David Anderson, director of the BOINC software platform that enables volunteer computing projects like SETI@home and climateprediction.net, pointed out that the process people go through to decide which project to donate their computer time to leads them to learn much about current scientific research and the process of science. Charles Loomis from the Enabling Grids for E-sciencE project discussed the use of EGEE and the BOINC software platform for drug discovery and medical imaging research.
“There are significant challenges ahead to sustain our planet for our great and great-great grandchildren,” Pordes concluded at the end of her presentation. “By stimulating cooperation, thinking and innovation, cyberinfrastructure and grid technologies are tools that will help us meet those challenges.”