ATHENA SEMINAR SERIES: Optical Networks on Chip – EDA Research Achievements and Perspective on Optical Computing
This event has passed.
Thursday, September 8, 2022 - 15:30 to 16:30
The NSF AI Institute for Edge Computing Leveraging Next Generation Networks (Athena) presents the next in our seminar series by Dr. Ulf Schlichtmann of Technical University of Munich, titled “EDA Research Achievements and Perspective on Optical Computing” on Thursday, September 8, 2022, from 3:30-4:30pm EST at Duke University Wilkinson building room 321 and via Zoom. Light refreshments will be served.
PRESENTER: Dr. Ulf Schlichtmann of Technical University of Munich
Abstract: Optical Networks on Chip (ONoCs) are a promising technology to resolve some issues which are increasingly plaguing traditional electrical NoCs. Excessive power consumption is chief among these issues. As researchers started looking into architectural options for ONoCs, it soon became apparent that Electronic Design Automation (EDA) would be very beneficial to improve such architectures and especially their physical implementation, e.g. due to the complexity involved. This is true already on a netlist level, but even more so once physical design is considered. Thus, since more than 10 years, researchers have started working on EDA approaches for the design of ONoCs.
At the same time, optical computing is seeing increasing attention. Especially the rise of new computing loads such as neural network accelerators has drawn attention to using light as a computation medium. On the one hand, data propagation inside optical computing units happens at the speed of light. On the other hand, properties of light signals such as phases can be modulated efficiently with a very low power consumption to implement computing functions. However, optical components are very sensitive to manufacturing and environmental variations.
I will review some of our research on EDA for ONoCs, with an emphasis on Wavelength-Routed ONoCs (WRONoCs) and will discuss current challenges in further improving EDA results. This will be followed by a look at opportunities how EDA research can further improve ONoC architectures. Opportunities exist especially in simultaneously considering multiple design aspects. I will look at how in optical computing for neural networks variations and thermal effects can be countered to ensure accuracy.
Finally, I will briefly consider the question how we can potentially combine the ONoC and the optical computing research synergistically from an EDA perspective going forward, which may lead to novel high-performance and energy-efficient all-optical computing solutions.
Bio: Ulf Schlichtmann holds a doctorate in electrical engineering from Technical University of Munich (TUM), as well as a technology business degree. He spent about 10 years in the semiconductor industry (Siemens, Infineon) in various engineering, management, and executive positions, working on design automation, design libraries, IP reuse, and product development.
In 2003, he joined TUM as professor and head of the Chair of Electronic Design Automation. From 2007-2013 he served as Dean and Vice Dean of TUM’s Department of Electrical and Computer Engineering (ECE). Since 2013, he serves as Associate Dean of Studies for International Programs, overseeing both the Department’s educational programs in Singapore and English language programs in Munich. Since 2016, Ulf is an elected member of TUM’s Academic Senate as well as the TUM Board of Trustees. He is a member of the German National Academy of Science and Engineering, and a board member of Germany’s edacentrum. Ulf serves on various private and public advisory boards, among them the advisory board of TUM’s Institute for Advanced Studies and the Bavarian School of Public Policy. Since 2021, he also serves as CEO of TUM’s Singapore research center, TUMCREATE.
Ulf’s current research interests include computer-aided design of electronic circuits and systems, with an emphasis on designing reliable and robust systems. His research especially addresses emerging technologies, such as microfluidic biochips, photonic interconnects, optical computing, and memristor-based neural network accelerators.
