Intelligent Computing Systems for Extreme-Efficiency and Security

Speaker: Raghavendra Pothukuchi
Affiliation: University of Illinois

Abstract:  We are at an exciting time when computing is becoming personalized and ubiquitous. Vital to the continued success of computing are systems – from internet-of-things (IoT) devices to datacenters – that can deliver on many fronts like performance, energy use, reliability and security, with only limited resources to expend, like energy, storage and time. Building such systems is hard, particularly when the operating environment is becoming more dynamic, and systems are becoming heterogeneous. Unfortunately, we continue to build systems with ad hoc heuristic policies that are suboptimal, unsafe, and non-composable.

My vision is to develop a new generation of computing systems that deliver extreme efficiency, together with reliability and security. Each component in the computing system continuously senses its execution and configures itself using intelligent control derived from principled methods like formal control and machine learning. In my talk, I will describe the techniques and prototypes I developed so far, which cover multiple system layers and heterogeneous hardware, and present the remarkable benefits of systems built with intelligent control.

Biography: Raghavendra (Raghav) Pothukuchi is a PhD candidate with Prof. Josep Torrellas at the University of Illinois. His research is on building intelligent systems for extreme-efficiency and security. He has interdisciplinary interests in computer system architecture, OS and runtimes, distributed systems, machine learning, formal control, and security. He is a winner of the W. J. Poppelbaum Award at Illinois for architecture design creativity, a Mavis Future Faculty Fellowship at Illinois, an ACM SRC competition, and was chosen as a rising star in computer architecture in 2018. He collaborates with AMD on modular control for heterogeneous systems, and this work resulted in a joint patent. He received his Masters in CS from Illinois in 2014 and worked at Nvidia before graduate school. He has a Bachelors (Hons.) from the Birla Institute of Technology & Science (BITS) Pilani, India where he was the university topper.

For more information, contact Prof. Danijela Cabric ()

Date/Time:
Date(s) - Feb 10, 2020
11:00 am - 12:30 pm

Location:
EE-IV Shannon Room #54-134
420 Westwood Plaza - 5th Flr., Los Angeles CA 90095