Research Overview

I am actively pursuing research in the following areas.

Stochastic Computing picture1
Extreme Energy Efficiency picture1
Exploiting Parallelism and Scalability picture1

Stochastic Computing explores approaches for dramatically improving energy efficiency by re-designing processors and applications from the ground up to exploit software- or hardware-based noise tolerance. Current projects include development and automation of tools and methodologies for stochastic software and hardware design and architecture.

I am exploring unconventional approaches for achieving extreme energy efficiency, especially in the context of ultra-low-power embedded systems and exascale systems. Research directions include mitigation and elimination of traditional scalability and energy efficiency bottlenecks through approximate computing, application-aware optimization, peak power management, and variation- and application-aware energy management.

Some of my research involves exploiting parallelism and scalability to enable new science and optimizing architectures, algorithms, and data structures to achieve orders of magnitude improvements in performance scalability and energy efficiency. Current projects include enabling novel sensor networks and accelerating CAD tools by exploiting parallelism and scalability afforded by massively-parallel processors.