President Clinton has appointed CRPC Director Ken Kennedy to co-chair a new committee that will advise the federal government on issues that impact high-end computing, information technology, and networking. Kennedy, a Noah Harding Professor of Computer Science at Rice University, was one of 20 representatives from academia, industry, and government to be invited to serve on the Advisory Committee on High Performance Computing and Communications, Information Technology, and the Next Generation Internet. CRPC External Advisory Committee members David Cooper of Lawrence Livermore National Laboratory and Steven J. Wallach of Hewlett-Packard's Convex Computer Corporation will also serve on the committee.
The committee will advise the White House through the Office of Science and Technology Policy (OSTP). Kennedy expects that it will "focus on federal programs of research investment in high-end computer, information, and communication technologies, like the Next Generation Internet, but will stay away from regulatory issues."
In addition to directing the CRPC and establishing many of its programs, Kennedy has made major contributions to the field of high performance computing. His research accomplishments with the CRPC include developing effective machine-independent parallel programming interfaces. He and CRPC collaborators proposed Fortran D, an extended version of Fortran that permits the specification of data distributions for arrays across the processors of a parallel machine. Kennedy directed a prototype compiler development effort at Rice that validated the concept. This effort led to the establishment, under his direction, of the High Performance Fortran Forum, a broad-based consortium to develop extensions to Fortran 90 aimed at high performance on parallel machines. The resulting standard for High Performance Fortran has found wide acceptance in the HPCC community.
Before leading the CRPC, Kennedy's pioneering efforts were recognized in the HPCC community and industry. While he was on sabbatical at IBM in 1978 and 1979, he began the development of one of the earliest and most successful automatic vectorization systems for Fortran. This project, which was continued with IBM support after his return to Rice, influenced the design of many commercial products. In the early 1980s, he began to extend his methods for vectorization to automatic parallelization. His work on the Rn programming environment led to the development of new algorithms for interprocedural data flow analysis. The approach used in the Rn environment was adopted in the Convex Applications Compiler, the first commercial compiler to use interprocedural optimization in Fortran. In the late 1980s, Kennedy and his group produced the ParaScope Editor, an advanced editor and interactive analysis system for parallelization of Fortran 77 that has been widely distributed and is the basis for at least one commercial product.
Kennedy has been a faculty member at Rice since 1971 and a Noah Harding Professor of Computer Science since 1985. He has been recognized as an outstanding teacher, receiving the George R. Brown Award for Superior Teaching in 1979. His 30th Ph.D. student will graduate this year. He is also a founding member of the W. M. Keck Center for Computational Biology at Rice University, a joint effort of Rice, Baylor College of Medicine, and the University of Houston.
In 1990, Kennedy was elected to the National Academy of Engineering. Currently, he is a Fellow of the American Association for the Advancement of Science, the Institute of Electrical and Electronics Engineers, and the Association for Computing Machinery. He won the W. Wallace McDowell Award for Contributions to Compiler Optimization and Leadership in Software Development for Parallel Computation in 1995. He is on the Board of Directors for Tera Computer Company (Seattle, WA) and AccessWare (Houston, TX), and is a consultant for Hewlett-Packard (Palo Alto, CA). He has also served as Artistic Advisory Vice President of the Houston Society for the Performing Arts since 1986.
|Other Issues of PCR||Back to PCR||CRPC Home Page|