Computer technology has witnessed three orders of magnitude growth in the last twenty years leading to substantial infiltration of computing platforms into systems we interact with every day. Two related scaling trends are primarily responsible for this spectacular growth: Moore’s law (exponential improvements in transistor density) and Dennard scaling (device scaling associated with improving switching speeds and power dissipations). These two technology trends have created the so-called “free lunch” of simultaneous improvements in transistor densities, switching speeds and power dissipation levels leading to about doubling of device densities per generation with 40% improvement in performance while keeping power dissipation the same. The global supply chain has also benefitted the semiconductor industry. The rise of fabless design companies and IP vendors implies that corporations do not necessarily have to own expensive fabs to develop and market their products.

The core theme of our research is investigating design methods for building fast, efficient and secure computer systems of the future. Two main trends will have a profound impact on computing technology in the coming years. One is the recent slowdown of Moore’s law and Dennard scaling making it highly difficult for us to continue making computers faster and energy efficient simultaneously. In addition, the applications are also exploding in diversity with the emergence of applications for sophisticated data analysis tasks such as optimized search, deep learning, relational analysis, probabilistic computation, and deductive reasoning. The computers of the future would use a mix of CMOS and non-CMOS devices, rely on radical non von-Neumann architectures and subject to a host of security and reliability related issues. Computers of the future would need to be introspective, adaptive and self-healing for delivering the desired computational power as efficiently and as securely as possible. Through our various projects we are investigating a myriad of different research problems which address such challenges. Please feel free to scan this webpage for detailed information about the specific research projects we are pursuing.

Education is a key objective of this group and we strive to train the next generation of academicians and industry professionals that will push the frontiers of technology in whatever they choose to do.