Skip Navigation
Staff Directory | Contact CISE | Contact Web Master | Site Map
National Science FoundationCISE - The Directorate for Computer and Information Science and Engineering
Graphic Line
Home | About CISE | Funding | Discoveries | News & Events | FAQs
Graphic Line
Header Graphic
Graphic

Divisions
Computing &
Communication
Foundations
  News
Bullet Computer &
Network Systems
  News
Information &
Intelligent Systems
  News
Bullet Shared
Cyberinfrastructure
  News
News and Events Links
Bullet CISE News Highlights
Bullet CISE Public Notices
Bullet Lectures
Bullet Workshops
Bullet CISE Event Calendar
Bullet Assistant Director's Presentations
Bullet Vacancies


CISE Distinguished Lecture Series

CISE Lectures

The Hot Design Challenge: Low Power Systems

Mary Jane Irwin

06/12/2001
2:00 pm - 3:00 am

4201 Wilson Blvd.
Stafford I 110
Arlington, VA 22230

Dr. Mary Jane Irwin is a distinguished Professor of CSE at Pennsylvania State University.

The explosive growth of mobile computing devices that at are typically constrained by battery life has brought hardware and software techniques for energy conservation into the spotlight. Irwin will address ways to improve energy consumption across the design hierarchy - from hardware optimizations to software optimization to alternative application solutions approaches.

After a brief review of the sources of energy consumption in CMOS circuits, Irwin will discuss a variety of innovative approaches to reducing system energy. Although circuit level optimizations like voltage and clock frequency scaling are critical for reducing energy consumption, switching activity is largely determined by the software running on the system. A number of compiler optimizations and their interactions with architectural design decisions will be discussed. Alternative applications approaches can also be considered. For example, the MPEG-4 video coding algorithm supports different quantization parameters, rate control, and spatial and temporal scalability options all of which impact energy consumption. The quantization parameter can be used to trade-off quality for communication energy. Rate control can be used to select the quantization parameter adaptively by setting a target bit rate. Irwin will conclude with a look at some future low power design challenges.

Dr. Irwin received her Ph.D. degree in computer science from the University of Illinois in 1977. Her current research and teaching interests include computer architecture, computer arithmetic, VLSI systems design, low power design, and electronic design automation. For her research contributions, Dr. Irwin was named Fellow of the IEEE in 1995 and Fellow of the ACM in 1996.

 

 

 

Graphic Space Graphic Space Graphic Space
Top
Bottom Corner Graphic Space Bottom Corner