Intel Research Seminar


DATE: Thursday, April 24, 2003
TIME: Noon - 1:30 pm
PLACE: Intel Seminar (417 S. Craig Street - 3rd Floor)

SPEAKER:
Padmanabhan (Babu) Pillai
University of Michigan

TITLE:
Energy Conservation and Adaptation in Mobile and Embedded Systems


ABSTRACT:
In recent years, there has been a rapid spread of non-traditional computing platforms, especially mobile and portable computing devices. As applications become increasingly sophisticated and processing power increases, the most serious limitation on these devices is the available battery life. This talk presents two approaches to conserving energy in mobile and embedded computing systems.

The first half of this talk will show how an operating system can exploit hardware features to change the processor operating frequency and voltage to reduce energy consumption. This is difficult in many embedded situations due to the strict timing issues involved in control or communication applications. A class of real-time dynamic voltage scaling (RT-DVS) algorithms are introduced that can provide dramatic energy savings while maintaining timing guarantees.

Although such mechanisms can improve energy efficiency of a system, there is no assurance that the limited energy available is expended on the most valuable tasks. The second half of this talk presents the Energy-Aware Quality of Service (EQoS) framework that can provide varying service quality to different applications based on task characteristics, value of results, and runtime requirements to maximize returns on available energy. Energy-based adaptation of the task set and service levels is formulated as a tractable minimization problem, with both optimal solutions and fast heuristics presented.

BIO:
Padmanabhan Pillai, also known as Babu, is currently finishing his Ph.D. in Computer Science and Engineering at the University of Michigan. He has received an M.S. degree from Univ. of Michigan, and B.S. from Carnegie Mellon University (ECE'96). Currently, he is affiliated with the Real-Time Computing Laboratory, and has focused his research in optimized operating systems for embedded devices, power management in real-time systems, energy-aware quality of service adaptation of applications, and the measurement and modeling of task energy consumption. He is interested in the management and implementation of ubiquitous computing devices, low-energy architectures, wireless networking, and novel OS service models.


For Further Seminar Info:
Contact Kim Kaan, 412-605-1203, or visit http://www.intel-research.net.

SDI Home: http://www.pdl.cmu.edu/SDI/