James Decker: Broadening NERSC's Mission

By James Decker, Deputy Director, Department of Energy, Office of Science

The NMFECC — NERSC’s predecessor — was the creation of Al Trivelpiece when he was the Director of Research for the Office of Fusion Energy, and of his Deputy, Bennet Miller. Several years later, when I directed that same office, since renamed the Division of Applied Plasma Physics, I assumed responsibility for managing Al and Bennet’s creation. When Al returned to the Department of Energy as Director of Energy Research,he selected me as his special assistant and gave me the task of studying how to provide the same kind of supercomputer support that fusion enjoyed to the other research programs in ER. At the same time, there was concern about the future of the supercomputer industry in the United States and the lack of access to supercomputers by the scientific community. I was selected by the President’s Science Advisor to chair an interagency committee charged with putting together a report recommending a U.S. program addressing both the needs of researchers and the state of the industry. The result of both of these tasks included a recommendation that the NMFECC role be expanded to include other areas of research.

We began with a pilot program that allocated 5 percent of NMFECC’s resources to other ER research programs.The pilot was an unqualified success. In fact, it was oversubscribed by a factor of 20 — so we changed the mission of the NMFECC from supporting fusion alone, to supporting all of ER’s research programs. At that time, it was the only supercomputer center dedicated to civilian scientific users; neither NSF nor DOD had similar centers. In 1990, the NMFECC was renamed NERSC, and when what later became the Office of Advanced Scientific Computing Research was formed, I was named its first director, once again managing NERSC in its broader role. I have continued my interest in its progress since becoming the Deputy Director of Energy Research in 1986.

Looking back over my nearly 25 years of involvement with NERSC, I think that it has been an outstanding success. It has made significant contributions to all of our research programs. Through ESnet, it has made a wide range of collaborations possible and continues to do so. It has also made substantial contributions to scientific computing and reliable high speed communications. In the future, I expect that the importance of computing, and of NERSC, to the Office of Science can only increase. Many of the problems we are interested in solving are of such complexity that they can only be addressed through scientific simulation, and NERSC must be a major player in developing the tools and techniques to solve them.

This article was published in the NERSC 25th Anniversary Brochure.