Monday, September 10, 2001. 9:00-10:00
D. Isaac (VP Science and Technology, IBM Research).
Influence of Technology Directions on System Architecture
computational performance of information systems has progressed over
the years in a manner similar to Moore's Law, or actually even faster.
This phenomenal growth in system performance will continue as we move
into the 21st century. However, there will likely be a shift in the
system components that are predominantly responsible for this continued
growth. Physical limits in the base technology will alter key components
and influence system architecture. For example, as conventional FET
transistor scaling approaches fundamental, atomic-scale limitations,
the emphasis shifts to implementing new Materials and devices. Integration
density is also beginning to be impacted by power density and cooling
limitations as well as overall system power usage. The primary metric
in system performance then shifts from computations per second to computations
per second per Watt. This metric will become critical for all systems
and not just battery-operated devices. Parallel architectures can begin
to address the power constraints, especially with concomitant advances
in system integration such as multi-core CPU chips, high-density/high-performance
packaging, and appropriate mixing of electronic and optical interconnections.
Dr. Randall D. Isaac is the Vice President, Science and Technology,
for the IBM Research Division. He has worldwide responsibility for the
Research Division's strategy in the areas of Physical Sciences and Technology,
including semiconductor, packaging, communications and display technologies.
He was formerly the Director of the newly formed IBM Austin Research
Laboratory in Austin, Texas. The focus of the lab is high-performance
microprocessor design. Prior to his current role, Dr. Isaac was a senior
manager in the Semiconductor Research and Development Center of the
IBM Microelectronics Division in Burlington, Vermont. In this capacity,
he was the project manager for the 64Mb DRAM development joint program
with Siemens and Toshiba. Dr. Isaac previously worked in IBM Research
in Yorktown as the Director of Silicon Technology. He also managed the
bipolar technology group and the silicon processing facility, and was
active in advanced silicon facility planning. Dr. Isaac received his
B.S. degree in physics from Wheaton College in Wheaton, Illinois in
1972 and his M.S. and Ph.D. degrees in physics from the University of
Illinois at Urbana-Champaign in 1974 and 1977, respectively. Dr. Isaac
joined IBM in 1977 at the IBM Thomas J. Watson Research Center at Yorktown.
Dr. Isaac is a Senior Member of IEEE and an American Physical Society
September 11, 2001. 9:00-10:00
Rattner (Intel Fellow and Director of Microprocessor Research Labs).
Electronics In The Internet Age.
In the last 50 years electronics has gone through revolutionary changes taking us from the first transistor in 1947 to a microprocessor with tens of millions of transistors today.
Over this period the advantages gained from Large Scale Integration (LSI) have improved performance, reduced cost, lowered power consumption and increased reliability of semiconductor devices. Costs have dropped by as much a 5 orders of magnitude bringing the transistor in an integrated circuit below 1/1000 cent. This revolution has enabled major advances in both the computer and communications industries. It has brought previously inconceivable computing resources to our desktops and our homes and connected these resources into the world wide web. When Gordon Moore plotted the curve showing that the number of transistors on a microchip would double every 18 months, no one had any idea it would become the law by which the computer industry lives. During 35 years, this law has been guiding the industry there have been a staggering number of breakthroughs. If the law is to hold for another 10 years we will face new challenges and some fundamental limits of physics. For example, as transistors shrink to dimensions measured by a few atoms quantum mechanical effects play a significant role in their behavior. This shift from micro-electronics to nano-electronics will require major changes in future designs. Personal computers and the Internet have already been the catalyst for significant social change. The continued increase of performance and reduction in cost of these resources will change social, political and commercial life in ways we are only just starting to understand.
This talk will give you a glimpse into an exciting future with new engineering, physics and social challenges.
Justin Rattner is an Intel Fellow and Director of Intel's Microprocessor Research Laboratory. His current R&D activities focus on future generation IA-32 and IA-64 microprocessor circuits, architectures and use models.
In December of 1996, Mr. Rattner was featured as Person of the Week by ABC World News for his visionary work on the Department of Energy ASCI Red System, the first computer on earth to achieve sustained performance of one trillion scientific calculations per second. In 1989, he was named Scientist of the Year by R&D Magazine for his leadership in parallel and distributed computer architecture. Mr. Rattner was recently honored as one of the Computing 200, the Computer Museum's project to recognize the 200 leaders having the greatest impact on the U.S. computer industry today.
Mr. Rattner joined Intel in 1973. He was named its first Principal Engineer in 1979 and its fourth corporate Fellow in 1988. Prior to joining Intel, Mr. Rattner held positions with Hewlett-Packard Company and Xerox Corporation. He received BS and MS degrees from Cornell University in Electrical Engineering and Computer Science in 1970 and 1972 respectively.
September 12, 2001. 9:00-10:00
Emer (Intel Fellow, Intel Architecture Group Director).
The Post-ultimate Alpha.
Joel S. Emer is an Intel Fellow based at the Massachusetts Microprocessor
Design Center, where he leads microarchitecture research efforts. Emer
joined Intel as part of a June 2001 agreement with Compaq Computer Corporation
that called for the transfer of microprocessor engineering and design
expertise to Intel.
to joining Intel, Emer was a Compaq Fellow and Director of Alpha Architecture
Research, where he led research efforts for future processors for Compaq's
64-bit family of servers. With 22 years of combined service to Compaq
and Digital Equipment Corporation, Emer has held various research and
advanced development positions investigating processor microarchitecture
designs and developing performance modeling and evaluation techniques.
Emer is recognized as one of the developers of the widely employed quantitative
approach to processor performance evaluation. More recently, he has
been recognized for his contributions in the advancement of simultaneous
multithreading technology. He holds nine patents with three pending.
He has also published more than 30 papers.
Emer received a bachelor's degree with honors in electrical engineering
in 1974, and his master's degree in 1975 -- both from Purdue University.
Emer earned a doctorate in electrical engineering from the University
of Illinois in 1979.