UCSD Jacobs School of Engineering Bay Area Alumni Reception, hosted by Sun Microsystems, Date and Time Goes Here

Speaker Bio

Photo of Tajana Rosing

Professor Tajana Rosing is an expert on optimizing the battery life, communication, and storage of portable electronics devices such as PDAs, cell phones, and laptops as well as sensors and other embedded systems. In addition, she has recently demonstrated large improvements in energy consumption and reliability of systems on a chip by performing joint optimization of power and topology design. At Hewlett-Packard Labs in Palo Alto, CA, from 1998 to 2004, Tajana Rosing led a team of researchers focused on low-power wireless media and embedded systems. She filed five patents last year alone on technologies ranging from wireless resource management in heterogeneous networks, to optimization of power consumption in I/O of embedded devices.

Rosing has devised algorithms that reduce the power consumption of small, portable, and inexpensive computing systems through the integration of intelligent hardware, software and wireless system design. For example, she demonstrated that large energy savings can be achieved by intelligent management of delivery of streaming video data to mobile nodes. After each burst of data is received, a power-hungry wireless cards can be switched to sleep mode while compressed data are being decoded. She wrote a chapter titled "Dynamic Management of Power Consumption" in Power Aware Computing, a book describing the trend towards embedded systems and systems-on-a-chip computing, in which power management has emerged as a focal point. Rosing is a leading researcher in the area of using information present in wireless systems to achieve more efficient system operation. For example, since a smart wireless server has access to information about a wireless device's characteristics and client needs in its environment, the server can schedule according to rules that maximize quality of service while minimizing power consumption. Rosing's server-level scheduling algorithm has demonstrated a significant increase in the number of clients supportable by a given server, with large power savings for each client.

footer bar
All Content Copyright 2011 Regents of the University of California. All rights reserved.
Official Web page of the University of California, San Diego.