7.29.05 -- In the latest issue of the high-performance computing news service HPCwire, San Diego Supercomputer Center (SDSC) director Francine Berman authors an Op-Ed article on the role that cyberinfrastructure research must play to maintain the competitiveness of U.S. science and engineering. Berman is affiliated with Calit2 and holds the High Performance Computing Endowed Chair in the Jacobs School's Computer Science and Engineering department. Her comments are re-printed below:
Recent articles in community publications have focused on the critical need for capable high performance computing (HPC) resources for the open academic community. Compelling reports from the National Research Council, PITAC (President's Information Technology Advisory Committee), the National Science Board, and others point to our current diminished ability to provide adequate computational and data management support for U.S. researchers, and the impact of insufficient technology capacity and capability on the loss of U.S. competitiveness and leadership.
As stated compellingly and increasingly, adequate capability and capacity in HPC is necessary, but it is not sufficient for leadership and competitiveness in science and engineering. Beyond the gear, concrete and strategic goals are critical to achieve competitiveness in science and engineering.
What do we want to accomplish as a nation in science and engineering? Competitiveness for many is reduced to an HPC "arms race" -- who has the top spots on the Top500 list? For others, competitiveness amounts to U.S. dominance in the science and engineering world, represented by the number of awards, prizes, and other recognitions for U.S. researchers. For still others, competitiveness is represented by what researchers and educators see as the diversion of a looming "perfect storm": decreasing funding for science and engineering in the U.S.; increasing outsourcing of people and ideas to Europe, Asia and elsewhere; and a decreasing number of students graduating in the sciences and engineering.
For any definition of competitiveness, the means to the end is a serious application of the Gretzky Rule: "Skate to where the puck will be." It is clear that we need concrete goals and a plan, timetable, and resources to achieve them. But what should our goals be? Which goals should have priority over others? How should we accomplish our goals? More funding is an easy answer, and indeed, nothing substantive can be done without resources. But leadership, concrete goals, and a strategic plan for achieving these goals ranks just as highly to ensure that funding is well spent and our efforts are successful.
So how can we apply the Gretzky rule to the going definitions of competitiveness?
The Gretzky Rule and Competitiveness in the HPC "Arms Race"
These days, competitiveness in high performance computing is commonly measured by ranking on the Top500 list. This approach is inadequate to really measure architectural innovation, robustness, or even performance on applications which do not resemble the LINPACK benchmarks; however, it is an easy measure and it has been effective in making the case for competitiveness beyond the scientific community. The current top spot on the list is occupied by Livermore's Blue Gene, however the emergence several years ago of the Japanese Earth Simulator (now at spot 4) provided a "wake up call" (Jack Dongarra from the University of Tennessee, Knoxville called it "computenik" in the New York Times) to the U.S.
The Earth Simulator provides a textbook application of the Gretzky Rule: Japan committed roughly 5 years and 500 million dollars to planning and executing the Earth Simulator, which stayed at the top spot on the Top500 list between June 2002 and June 2004. Careful planning, investment, and commitment enabled the Earth Simulator to create an impact that is still being felt in the U.S. and Europe.
So what did we learn about competitiveness from the Earth Simulator? A concrete goal achieved by strategic planning, commitment, and resources over an appropriate timeframe made this a reality.
The Gretzky Rule and Competitiveness in Science and Engineering Research
For most academics, competitiveness is measured by quality of results and track record through publications, and the most highly valued research and researchers are candidates for community prizes -- the Fields Medal (mathematics), the Turing Award (computer science), the Pulitzer Prize (literature), and of course, the Nobel Prize (various disciplines). The ultimate goal of competitiveness is leadership, and to achieve the kind of leadership recognized by community prizes, researchers must devote many years in an environment that supports creativity, innovation, deep thinking, and does not penalize the many false starts, wrong turns, and other building blocks that lead to our best and most important results.
To create an environment in which U.S. scientists and engineers are competitive involves developing an environment where the best, the brightest, and the most creative can work, and over the long periods of time required for fundamental advances. For many of today's scientists and engineers, infrastructure and professional support is decreasing in the university environment, and there is increasing difficulty in getting funded by federal agencies. (Currently the "hit rate" for computer science and engineering proposals at the National Science Foundation is 20 percent or less, i.e. only one in every five proposals is funded). In addition, increasing risk aversion in the funding environment penalizes bold, long-term, or unusual approaches.
Optimizing for competitiveness in science and engineering research mandates a different approach than the HPC "arms race" to the provision of high performance computational and data management infrastructure as well. Rather than optimizing for Top500 ranking, enabling HPC platforms for the researchers who need them must optimize for the support of real science and engineering applications. Data-intensive HPC applications, latency tolerant grid-friendly applications, latency-intolerant "traditional HPC" applications, etc., require a diverse set of capable and high-capacity HPC architectures to best support the diverse needs of the broad academic community. One size (architecture or site) does not fit all here. At the same time, we can't currently afford hundreds or perhaps even tens of these facilities -- economies of scale must be applied to optimize for adequate capacities and capabilities, as well as the costs of support, maintenance, and user service and training required to best leverage national-scale HPC resources for the broad community.
So how can we become more competitive in U.S. science and engineering research? Our research and education portfolio would benefit from the same approach we use to balance our personal investments. We should be investing in a strategic balance of short-term, long-term, high-risk, and low-risk endeavors. We should acknowledge that infrastructure enables new discovery but also incurs cost and requires stability. If the "puck" is leadership through a greater U.S. percentage of top prizes and high-impact results, we need to focus our resources on developing an environment where this can happen, and begin skating in that direction.
A Perfect Storm Looming
The outsourcing of research, education, service, and innovation is an increasing focus for discussion in the public and private sector. According to Science Resource Statistics, as of 2003, 22 percent of professional scientists and engineers did not have a B.A. or B.S. and only 9 percent held Ph.D.s and professional degrees. The number of doctorates awarded in science and engineering have generally been decreasing since 1998, and despite the fact that our kids are increasingly technology-savvy, as a society, our understanding of science and engineering is seriously limited. The National Science Board's 2004 Science and Engineering Indicators report states, "Many people do not seem to have a firm understanding of basic scientific facts and concepts. Experts in science communication encounter widespread misunderstanding of how science works."
For many of us in academia, the increasing competitiveness of our colleagues in Europe and Asia through committed funding programs and resources, the drop in support in the U.S. for research, education, and information infrastructure, and the increased outsourcing of technology innovation and service outside of the U.S. are creating a "perfect storm" that will batter U.S. leadership and competitiveness not just now, but over the next generation. Investment in maintaining and sustaining a competitive U.S. workforce in science, engineering, and technology is a long-term investment. It will require planning, commitment, and resources for our educational system, expansion of our training environments, and evolution of our cultural perceptions to recognize the critical role science and engineering play in driving key societal challenges such as better health, improved safety, a sustainable environment, etc.
If we have a concrete idea of where we want the puck to be, it's much easier to skate there. Setting strategic priorities and concrete goals, committing to providing the leadership, perseverance, and resources to meet those goals, and responsibly estimating the costs and the timeframes required to reach them are key to competitiveness and leadership. This is not rocket science, but without a more thoughtful and strategic approach, advances and new discoveries in rocket science and other disciplines will be much more difficult to achieve.
* This article is re-printed with permission from SDSC and HPCwire .