As momentum continues to build for this year's Supercomputing gathering (SC|05), nominations are being sought for this year's Sidney Fernbach Memorial Award.
Established in 1992, this prestigious award honors innovative use of high performance computing in problem solving. It is named for one of the true pioneers in the development and application of high performance computers for solving large-scale computational problems. Sid Fernbach remains a legend in the pages of HPC history, so his story is worthy of retelling – even for those who have prior or first-hand knowledge.
Born in Philadelphia, educated at nearby Temple University and trained as a physicist, Fernbach received a Ph.D. from Berkeley in 1952. Soon after he started on what was to be a storied career at the Lawrence Livermore National Laboratory. By 1955, he was named head of the lab's computation department, a position he would hold for the next two decades.
It was during this era, recalled Fernbach's friend and associate Alfred E. Brenner, “the modern electronic computer grew from embryo to adolescence.” Beginning with the acquisition of the computer known as Univac 1 in April 1953, Fernbach remained front-and-center regarding the procurement of subsequent computers and their ever-expanding capabilities.
Fernbach understood from his early days at Livermore that the operation needed increasingly potent computing power to deliver its designs. To this end, Fernbach forged lasting relationships with other prime influencers of that time, such as Seymour Cray. Their relationship started when Cray was at Control Data Corp. before founding Cray Research. For example, Fernbach's sharp insights, unwavering dedication and keen understanding of computational science led to convincing the Atomic Energy Commission (later known as the Department of Energy) to approve multi-million dollar supercomputer acquisitions.
Fernbach also headed the lab's theoretical division from 1958 to 1968 before moving on to serve as deputy associate director of scientific support from 1975 to 1979, when he left the lab.
Along the way Fernbach compiled a lengthy list of accomplishments including founding and chairing the IEEE's subcommittee on scientific supercomputing. In this capacity, he played a pivotal role in shaping government policy as it related to supercomputing supporting the well being of the United States' economy and security.
He served as a board member and consultant in the 1980s. At the age of 73, Fernbach died quietly in his sleep on Feb. 15, 1991, at his house in Alamo, Calif.
Nominations are being solicited from the conference committees of SC05 for this year's Fernbach candidates. The award and $2,000 honorarium will be given at SC `05 in Seattle this November.
For more information on the award, go to
http://newton.computer.org/awards2.nsf/nominationform?openform&code=fernbach or http://sc05.supercomputing.org/programs/awards_prizes.php. The deadline for nominations is July 31. Nomination forms and submission instructions can be found at http://www.computer.org/awards/.
The award is given annually by the IEEE Computer Society to an individual for “an outstanding contribution in the application of high performance computers using innovative approaches.” The list of past Fernbach award recipients reads like a who's who of an HPC history book. In a way, it's ironic that Fernbach would have an award named after him. After all, he eschewed most opportunities to be honored, his colleague Brenner said.
“Sid's most lovable attribute was an independent and a no-nonsense attitude toward getting the job done,” wrote Brenner and former Supercomputing Research Center colleagues John P. Riganati and Paul B. Schneck, in an obituary they co-authored. “He would not tolerate bureaucracy and when discussing an issue, he would quickly get down to the core of the problem. This characteristic earned him the reputation of a no-nonsense sage who gets things done.”
Fernbach's legacy is being carried on by the winners of his annual memorial. Last year, Marsha J. Berger became the first woman to be added to the list of distinguished members, which includes Jack J. Dongarra, Robert J. Harrison, 2001, Stephen W. Attaway, Michael Norman, Philip Colella, Charbel Farhat, Gary A. Glatzmaier, Paul Woodward, Charles Peskin and David H. Bailey.
In 2004, Berger, a professor at New York University's Courant Institute of Mathematical Sciences was honored “for her many contributions and enormous influence to computational fluid dynamics, including adaptive mesh refinement methods, Cartesian grid methods and practical mathematical algorithms for solving significantly heretofore intractable problems.”
“I was especially pleased to be `singled out' for the Fernbach award,” she said. “Since the award committee is very knowledgeable in this field, it is really an award by your peers. It's especially gratifying to have your work recognized by people who really understand it.”
“I didn't realize I was the first woman to win the award,” said Berger, who currently is on leave from NYU. “Since the award, my line of research for the last few years has pretty much continued. I am just finishing two years — one on sabbatical, one on leave — at NASA Ames, working with my long-term collaborator Michael Aftosmis. We are developing Cartesian embedded-boundary methods for flows in complex geometries, such as flow around the space shuttle, which is a very complicated geometry.”
The previous year, Dongarra, representing the Computer Science and Mathematics Division of the Department of Energy's Oak Ridge National Laboratory, won “for outstanding and sustained contributions to the area of mathematical software, most particularly in the areas of communication and numerical libraries and performance benchmarks for high performance computing.”
Now at the University of Tennessee's Innovative Computing Laboratory, Dongarra said, “I felt very honored to receive the Sid Fernbach Award for Computational Science in 2003. I had the pleasure of working with Sid on a couple of occasions and have great respect and appreciation for his contributions to our field of computational science.”
In 2002, Harrison, a quantum chemist and principal architect of the Northwest Computational Chemistry Software, was named “for developing a computational chemistry software package for applications development, by integrating fundamental algorithm research, novel ideas in computer science, and scalability, while delivering unprecedented modeling capabilities for chemistry applications.”
In 2000, Attaway, a distinguished member of the technical staff at Sandia National Laboratories, was chosen “for pioneering advances in methods for modeling transient dynamics phenomena, enabling simulations of unprecedented scale and fidelity.”
In 1999, Norman, a professor of astronomy at the University of Illinois at Urbana-Champaign and a researcher with NCSA and the Alliance Cosmology team, received the award “for his leading edge research in applying parallel computing to challenge grand problems in astrophysics and cosmology.”
In 1998, Colella, a mathematician and leader of the Applied Numerical Algorithms Group at the National Energy Research Scientific Computing Center, was chosen “for fundamental contributions to the development of software methodologies used to solve numerical partial differential equations, and their application to substantially expand our understanding of shock physics and other fluid dynamics problem.”
In 1997, Farhat, a professor and former chair of Aerospace Engineering Sciences and former director of the Center for Aerospace Structures at the University of Colorado at Boulder, was selected “for outstanding contributions to the development of parallel numerical algorithms and parallel software packages that have helped the mechanical engineering world to embrace parallel processing technology.”
“I was surprised to win, partly because all the previous winners were scientists rather than engineers — and I was not that well connected with the IEEE in those days,” said Farhat. “The award has affected my work in the sense that it added to its visibility.”
In 1996, ESS co-investigator Glatzmaier from Los Alamos was honored “for innovative computational numerical methods to perform the first realistic computer simulations of the Earth's geodynamo and its resultant time-dependent magnetic field.”
In 1995, University of Minnesota physicist Woodward was the award winner. His contribution focused on “developing new algorithmic techniques in fluid dynamics. He was further cited for his “relentless and innovative pursuit of the hardware and software capabilities to carry out and visualize in real-time the largest turbulence simulations” have stood the test of time.”
In 1994, Peskin, a scientist at New York University's Courant Institute of Mathematical Sciences, was honored “for innovative application of mathematical modeling methods to important practical research questions in blood flow and the heart that has for more than 15 years pushed forward the leading edge of computational capability and helped to develop supercomputing technology as a valuable tool for improving the quality of human life.”
In 1993 Bailey was picked “for contributions to numerical computational science, including innovative algorithms for FFT's, matrix multiply and multiple precision arithmetic on vector computer architecture.”
This year's winner will have an impeccable legacy to uphold, being recognized for working at the highest levels of proficiency that typified Fernbach's life and work.
Peter Meade is director of publishing operations at Tabor Communications Inc., the publisher of HPCwire. He would like to thank Alfred E. Brenner for his time, assistance and gracious cooperation in preparing this story.