Shanbhag receives 2010 Richard Newton GSRC Industrial Impact Award
CSL researcher Naresh Shanbhag recently received the Richard A. Newton GSRC Industrial Impact Award from the Gigascale Systems Research Center (GSRC), one of the six national centers funded under the Focus Center Research Program (FCRP) established by DARPA and SRC. Named after one of the founders of GSRC and a former Dean of College of Engineering at the University of California at Berkeley, the award honors GSRC principal investigators whose projects and/or concepts, initiated at least 5 years ago, have had substantial industrial impact.
The work for which he received this award began in the late 1990s. Specifically, he received the award for his work described in a series of papers on the concept of Algorithmic Noise Tolerance (ANT). In his research, Shanbhag came up with a novel approach to design energy-efficient and robust circuit and systems. Previously, circuits and architectures were designed to be error-free. Shanbhag proposed permitting circuits to make errors and correcting them in order to save power and enhance reliability much as is done in communication links. This would allow for less costly manufacturing of the integrated circuits in nanoscale process technologies.
“When I first proposed the idea that circuits should be permitted to make errors, there was a substantial resistance from the community,” said Shanbhag, a professor of electrical and computer engineering. “Today, error-resiliency is an accepted design approach in both industry and academe. So, for me, this award means a lot because recognizes the originality of the idea and its impact.”
In essence, Shanbhag’s work behind this award was on designing reliable and low-power computing systems in nanoscale process technologies. As circuits have been reduced to the nanoscale, problems with power and reliability have emerged as key areas of concern in designing chips. As size has reduced, it has become increasingly difficult to maintain error free circuits.
The work, which began in the late 1990s, continues to be a major focus of Shanbhag’s research. Working with his colleagues at Illinois (in particular Professors Douglas L. Jones, Andrew Carl Singer, and Rakesh Kumar) and around the country, Shanbhag is developing the concept of stochastic computing that builds upon the principles of ANT. “One of my major activities is to develop the foundation, that is, the principles of stochastic computing,” he said. “The second aspect of this effort is to develop practical architectures and implementations—very large scale integrated circuits or complex systems-on-a chip.”
In terms of applications for this type of system, Shanbhag sees biomedical as a particularly apt field for his type of research. “You want your system to be robust, and if they are portable you want the batteries to last a very long time,” he explained. “So biomedical could be a great application area. Many alternative/green energy sources tend to be stochastic in nature. Therefore, stochastic computing systems can be a natural fit.”
Shanbhag has been affiliated with GSRC since 2003. Since 2006 he has been leading a research theme in GSRC called Alternative Computation Models. Dr. Shanbhag became an IEEE Fellow in 2006, received the 2006 IEEE Journal of Solid-State Circuits Best Paper Award, the 2001 IEEE Transactions on VLSI Best Paper Award, the 1999 IEEE Leon K. Kirchmayer Best Paper Award, the 1999 Xerox Faculty Award, the Distinguished Lecturership from the IEEE Circuits and Systems Society in 1997, the National Science Foundation CAREER Award in 1996, and the 1994 Darlington Best Paper Award from the IEEE Circuits and Systems Society. He is a 1996 recipient of an NSF CAREER Award.