Argonne Wins R&D 100 Awards

Advances in technology ranging from help for victims of Parkinson's disease and epilepsy to more efficient combustion in industrial furnaces are likely with award-winning research at the U.S. Department of Energy's Argonne National Laboratory and its partners. Argonne's research accomplishments have won four of the prestigious R&D 100 Awards, given to the world's top 100 scientific and technological innovations. This is the second year in a row that Argonne has won four R&D 100 Awards. This year's awards bring Argonne's total to 90 of the awards since R&D magazine began presenting them in 1964. Argonne director Bob Rosner congratulated the winners, saying, "I am thrilled that Argonne staff members have won four more of these prestigious awards. Winning such awards attests to the high quality of research at Argonne and to the caliber of our staff.” “These awards demonstrate that DOE scientists and researchers are hard at work developing the technologies of the future,” said Secretary of Energy Samuel W. Bodman. “In the past, breakthroughs like these have played an important role in both our economic and national security.” One of this year's winners from Argonne is MPICH2, software that enables scientists to write parallel programs that run efficiently on all major computer systems, from parallel processors to laptops. MPICH2 is a high-performance, portable implementation of community standards for the message-passing model of parallel computation. Parallel computation requires many computers to work together on large-scale problems by quickly distributing the mathematical workload. The new software's layered architecture permits computer vendors and researchers to customize its lower layers for particular proprietary networks while using its portable upper layers to provide compliance with computer community standards and state-of-the-art computational algorithms. The software, developed at Argonne by William Gropp, Ewing Lusk, Robert Ross, Rajeev Thakur and Brian Toonen, enables application developers to run the same code on a wide variety of platforms, from laptops and workstations, through clusters of computers that can be assembled from off-the-shelf components, to the largest and fastest parallel computers in the world. Applications include materials science, combustion simulation, astrophysics, climate modeling and bioinformatics. MPICH2 is the first freely available open-source implementation of the MPI-2 international message-passing standard, and both users and vendors have been quick to adopt it. Companies such as Pratt and Whitney are using MPICH2 to design aircraft engines, and the software is also widely used in scientific applications. A new epilepsy modeling program from a neuroscience group at the University of Chicago was one of the first applications to take advantage of the new remote memory access functionality in MPI-2. In addition to the team listed above, significant contributors to the project include David Ashton at Argonne, Ralph Butler at Middle Tennessee State University, and Anthony Chan at the University of Chicago.