CIS PhD student Khadka receives A. Richard Newton Young Student Fellow Award
CIS first-year PhD student Shreeya Khadka was awarded the A. Richard Newton Young Student Fellow Award in April 2016. This award was set up to honor the memory of A. Richard Newton, a pioneer in electronic design automation and integrated circuit design.
The fellowship encourages and supports students at the beginning of their research career in Electronic Design Automation and Embedded Systems by providing them financial support to attend Design Automation Conference (DAC) and present their work. DAC is the premier conference devoted to the design and automation of electronic systems (EDA), embedded systems and software (ESS), and intellectual property (IP). With these awards, Shreeya Khadka will have the opportunity to participate in numerous activities during the conference, including meetings with design automation luminaries, attendance at technical sessions and exhibits.
Shreeya's research focus is the design exploration of self-aware adaptive architectures. In her current project, she is examining the use by computing systems of machine learning algorithm and control theory to self-monitor, analyze and autonomously adapt to the application execution.
At DAC she will introduce a neural network based predictive routing algorithm for on-chip networks which uses anticipated global network state and congestion information to efficiently route network traffic. The core of the algorithm is a multi-layer neural network machine learning approach where the inputs are level of occupancy of virtual channels, average latency for a particular router to be selected for route computation, the probability of virtual channel allocation, and the probability of winning switch arbitration at the crossbar. The algorithm lends itself to both node routing and source routing. In most test cases, the proposed approach outperforms current deterministic and adaptive routing techniques in terms of latency and throughput. The hardware overhead for supporting the new routing algorithm is minimal.