Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
9 Summary Doyle D. Knight Consortium for Scientific Computing The John con Neumann Nai'onal Supercomputer Center These proceedings of the symposium on supercomputers have focused on two major topics. First, Jack Worlton and Steve Chen described the evolution of supercomputer technology over the past 20 years and projected future trends in improved processor performance, increased memory, and rapidly expanding parallelism. They also emphasized the role of algorithm development, noting that improvements in performance associated with the development of new algorithms are increasingly important. Second, three presentations focused on current applications of su- percomputers in industry. Beverly Eccles (Abbott Laboratories), Clifford Perry (Eastman Kodak Company), and Larry Tester (Apple Computer, Inc.) provided a series of examples of applications of supercomputers in their corporations. Into key points were emphasized: · Supercomputers provide the opportunity to design new products, ranging from film emulsions to computer keyboards, with a degree of accuracy heretofore unachievable with conventional computers and at a cost oftentimes far lower than that associated with experimental methods (e.g., development of prototypes). · Supercomputers can improve the productivity of designers by sig- nificantly reducing the time required to evaluate a new idea. The rapid feedback of results enhances creativity. These three researchers also addressed the issue of integration of super- computer technology into industry. Four important points were stressed: 101
102 DOYLE D. KNIGHT · Industry must first recognize the potential benefits of supercom- puter technology in research and development. · A core of supercomputer "evangelists" must be established initially within a corporation. This core group people who are dedicated to using supercomputing as well as to explaining and communicating its advantages to other scientists in industry will provide the leadership and incentive for the adoption of the technology by the larger group. · "Bottom-up" planning is necessary for successful incorporation of supercomputers into industry. The current computer "habits" of researchers and designers must first be understood before any major changes can be implemented. · Close collaboration between academia and industry is needed to provide improved software tools and training for students, who will become the researchers in the industry of tomorrow.