Distributed Computing on the Global Network Computer:
A Vision
January 4 Gul Agha University of Illinois at Urbana-Champaign 8:00am - 9:30am Jade Room
Concurrent computing technologies are advancing rapidly. At one end of the spectrum, hardware technology is making multiprocessor workstations feasible. At the other, emerging information highways allow one to view the nodes connected to the internet as a global distributed computer. To effectively use the emerging technological infrastructure, parallelism and distribution must be understood in fundamental ways. In particular, the interrelated issues of naming, resource management, coordination, mobility, and availability are critical. I will discuss the current trends in programming paradigms, such as web browsers, agents, and reflective architectures. I will also discuss the state of the art in programming languages _ such as Java and Rosette _ which are in use in the real world. Finally, I will provide my assessment of the technological requirements for software to enable effective use of the emerging global network computer.
Professor Gul A. Agha is Director of the Open Systems Laboratory at the University of Illinois at Urbana-Champaign. His widely cited book, "Actors: A Model of Concurrent Computing in Distributed Systems", (MIT Press, 1986) has provided a basis for a number of research projects in concurrent computing. Dr. Agha is the Editor-in-Chief of IEEE Parallel and Distributed Technology. He is also Associate Editor of the journals ACM Computing Surveys and Theory and Practice of Object Systems. Dr. Agha is an ACM International Lecturer, a past recepient of the Incentives for Excellence Award from Digital Equipment Corporation, and of the Naval Young Investigator Award from the US Office of Naval Research. He was named a Fellow in the University of Illinois Center for Advanced Study in 1992. Dr. Agha has given over eighty invited lectures at universities, industrial laboratories, and international conferences in the past five years. He has published research papers in a number of areas related to distributed systems, including concurrent programming languages and their compilation, fault-tolerance, distributed real-time systems, mathematical foundations of concurrent programming, parallel algorithms and visual programming.
PLENARY 1 The Economics of Attention and the Arts
of Information
January 4 Richard A. Lanham President, Rhetorica, Inc. 1:00pm - 1:50pm Haku Room
Economics, in the classic definition, studies the allocation of scarce resources. In an information economy the scarce resource is not information itself - we're drowning in that - but the human attention which gives it meaning and direction. I begin by discussing the need for such an exonomics of attention and the obstacles, professional and political, which presently obstruct its formation. I then illustrate, in a multimedia presentation, how the economics of attention, though seldom treated in the social sciences, has constituted a main subject of the visual arts in the twentieth century. Next, I illustrate how digital videographics has both subsumed and extended this preoccupation with information and how we attend to it. I conclude by suggesting that these "arts of information," both conventional and digital, point toward a more interactive theory of communication than that to which an industrial "goods" economy has accustomed us.
Professor Lanham is the president of Rhetorica, Inc., a Los Angeles consulting and media production company. Dr. Lanham was educated at the Sidwell Friends School in Washington, D.C., and at Yale University, from which he holds the A.B., M.A., and Ph.D. degrees. He began his teaching career at Dartmouth and from 1965 to 1994 taught in the English Department at UCLA, where he is now Professor Emeritus. He has been a Senior NEH Fellow, a Senior Fellow in the Society for the Humanities at Cornell, a Guggenheim Fellow, and most recently, Norman Freehling Visiting Professor at the Institute for the Humanities at University of Michigan, and the 1994 International Scholar at the George Eastman House in Rochester, N.Y. He is the author of The Motives of Eloquence, Literacy and the Survival of Humanism, and eight other books of literary criticism and prose stylistics. His latest book, The Electronic Word: Democracy, Technology, and the Arts, was published, in both printed and electronic form, by the University of Chicago Press in December 1993. He is now developing a CD-ROM introduction to multimedia for Calliope Media.
PLENARY 2 Molecular Manufacturing: Bringing Software
Economics to the World of Things
January 5 K. Eric Drexler Institute for Molecular Manufacturing 1:00pm - 1:50pm Haku Room
In the world of information, digital technologies have made copying fast, cheap, and perfect, quite independent of the cost or complexity of the content. What if the same were to happen in the world of matter? The production cost of a ton of terabyte RAM chips would be about the same as the production cost of a ton of steel. Design costs would matter, production costs wouldn't.
By treating atoms as discrete, bit-like objects, molecular manufacturing will bring a digital revolution to the production of material objects. Working at the resolution limit of matter, it will enable the ultimate in minaturization and performance. By starting with cheap, abundant components _ molecules _ and processing them with small, high-frequency, high-productivity machines, it will make products inexpensive. Design calculations indicate that a mature technology of this sort can produce air-cooled desktop computers that each execute more instructions per second than all of the semiconductor CPUs in the world today combined.
Research programs in chemistry, molecular biology, and scanning probe microscopy are laying the foundation for a technology of molecular machine systems. Focused efforts today are centered in Japan, sponsored by STA and MITI; the US has a strong position in the basic technologies. It's time we paid more attention.
Eric Drexler is a researcher concerned with emerging technologies and their consequences for the future. This interest led him to initiate studies in the field of molecular nanotechnology - an anticipated technology based on the molecular machines able to build objects to complex atomic specifications.
Dr. Drexler received an S.B. degree from M.I.T. in Interdisciplinary Science, an S.M. degree from M.I.T. in Engineering (while a National Science Foundation Graduate Fellow), and a Ph.D. from M.I.T. in Molecular Nanotechnology. He lectures widely, and has published papers and articles for periodicals ranging from Smithsonian to the CoEvolution Quarterly to the Proceedings of the National Academy of Sciences. Formerly a Research Affiliate of the M.I.T. Space Systems Laboratory and the M.I.T. Artificial Intelligence Laboratory, he is currently a Research Fellow of the Institute for Molecular Manufacturing. He serves on the Board of Directors of the National Space Society and is member of the American Vacuum Society, the AAAS, and the American Chemical Society.
PLENARY 3 The Microprocessor and its Performance in
the Year 2000
January 6 Yale N. Patt University of Michigan, Ann Arbor 1:00pm - 1:50pm Haku Room
The marketplace continues to demand more and more performance from the computer systems we deliver. The solid state people are helping by promising 100 million transistors on a single chip by the year 2000. With that many transistors available, what will the microprocessor of the year 2000 look like? This talk will explore some of the possibilities, and examine the critical limitations to exploiting the capability suggested by all thos transistors. Furthermore, we will discuss what is needed from hardware people, compiler people, and algorithm people if we are to overcome those limitations, and satisfy the continually increasing demands of the marketplace.
Yale N. Patt is Professor of Electrical Engineering and Computer Science at the University of Michigan, Ann Arbor, where he teaches computer architecture and directs the research of twelve PhD students in high performance computer implementation. He is also developing, with support from Intel, AT&T/GIS, Digital Equipment Corporation, Hewlett-Packard, Motorola, and Tektronics, an experimental research laboratory for students to develop and measure the performance of real code executing on real systems. In addition he consults extensively in industry on problems related to implementing high performance computer systems. His clients include Digital Equipment Corporation (since 1977) and AT&T/GIS Division (since 1986).
Dr. Patt has published more than 100 articles in the computer science and engineering literature, has lectured at more than 50 universities, and regularly teaches on-site two-week courses on computer architecture and implementation to practicing engineers, and shorter courses to management personnel on the implications of various advances in computer technology. He particularly enjoys giving this HICSS tutorial because it provides an opportunity to explain the hot topics of computer architecture to an audience unfettered with too much conventional wisdom.
Dr. Patt earned his B.S. degree from Northeastern University and the M.S. and PhD from Stanford University, all in electrical engineering. He is the recipient of the 1995 IEEE Emanuel R. Piore Award. He is a Fellow of the IEEE.
HICSS-29 HOMEPAGE
HICSS HOMEPAGE