|Volume 7, Issue 1 -
Noah Harding Professor, Computational and Applied Mathematics, Rice University; Member of the Executive Committee and Director of the Parallel Optimization Group, Center for Research on Parallel Computation
John Dennis' research is on practical methods for optimization, with particular interest in parallel methods for nonlinear optimization of systems described by coupled nonlinear simulations. His work in multidisciplinary optimization has already placed Rice and the CRPC at an advantage in computational engineering. By using parallel computers to solve optimization problems in engineering, Dennis has combined two separate areas in which Rice and CRPC faculty have tremendous experience.
"All of us in the CRPC Optimization Project are certain that optimization/ decision support has a central role in computational engineering," Dennis notes. This role is realized through many forms. "We are developing these optimization tools, including computational differentiation, in many collaborative efforts," he said. "I am finishing the first semester of a new course based on building optimization models of engineering case studies and solving the problems using software obtained from the Internet. In addition, we are making a serious effort to keep our project on track by communicating with industry."
Regarding the future of computational engineering, Dennis says that it is "more than just the increased use of computers in engineering. Computers are going to be used more extensively in every job, but we don't go around talking about computational lawyers or computational chefs. The difference is that engineering is going to be at a completely different level as a result of the HPCC infrastructure -- this is true for software as much as hardware. The obvious advantage will be in testing proposed solutions to engineering problems by using high-performance computing simulations. The key will not just be that more alternatives can be tested, but rather that alternatives can be simulated whose testing was not feasible before either because of the expense (helicopter design) or the danger (automatically controlled highways) of prototyping. Doing engineering in this environment with the freedom to ask new 'what if?' questions is going to lead to a new golden age of engineering creativity. This environment for doing engineering will also lead to quicker, better, cheaper product development. This is the reason it is crucial to our economy that we develop the HPCC infrastructure and teach our engineering students to exploit it."
Dennis sees the CRPC as having had a primary role in the development of computational engineering, ever since the center's founding in 1989. "CRPC has been fundamentally a computational engineering project from the beginning," he said. "Look at our mission: to make parallel computers truly usable. Our mission is exactly on target, as are our educational programs. Our role is to speed the process of developing the hardware and software and synthesizing the environment that will make computational engineering happen faster and in a way that will justify the tax dollars used to support CRPC.
Among his honors, Dennis is editor-in-chief and founder of the SIAM Journal for Optimization, an advisory editor of Mathematics of Operations Research, and a past co-editor of Mathematical Programming. He has served as chair of the SIAM Activity Group for Optimization and has served two terms on the SIAM Council. Currently, Dennis chairs the Mathematical Programming Society. Dennis has done extensive consulting for United States industry, has been a Fulbright Lecturer to Argentina, and has published his algorithm research in several widely disseminated software journals. He has given many short courses and featured addresses at international conferences. His textbook, co- authored with R.B. Schnabel, was published by Prentice-Hall and in Russian translation by Mir.
While at Rice, he has served as department chair in computer science and in computational and applied mathematics. Dennis has directed 30 Ph.D. theses and his students hold positions in industry, government, and academic departments of business, mathematics, computer science, and applied mathematics.
Table of Contents