January 11, 2013
Interschool Lab, 750 CEPSR
Speaker: Dr. Trent McConaghy, CTO (Solido Design Automation)
How does one handle PVT & statistical variation when designing analog and memory integrated circuits (ICs)? How can automation and machine learning aid analog and memory circuit design, in ways that designers might accept? What are the near-term and longer-term research opportunities related to analog circuits, machine learning, and the adjacent field of neuroscience? This talk addresses these questions.
First, this talk reviews research from the frontiers of computer-aided design (CAD) tools for custom ICs that is designed for, and used by, industrial custom IC designers at bleeding-edge process nodes; subtopics include high-sigma Monte Carlo analysis, fast PVT analysis, template-free symbolic modeling, and automation-aided trustworthy analog topology exploration. Then, this talk reviews near-term research opportunities on designing & verifying extremely large scale AMS and memory systems, under extreme variation. Finally, it reviews the longer-term research opportunities of designing at scale under radical IC device changes, and leveraging analog design and machine learning in big data / neuroscience applications.
Trent McConaghy is co-founder and CTO of Solido Design Automation, which provides variation-aware design tools for custom circuit designers at major semiconductor vendors & fabs including Nvidia, Qualcomm, and Huawei. He was a co-founder and Chief Scientist of Analog Design Automation, which Synopsys acquired in 2004. Prior to that, he did intelligent systems research for the Canadian Department of National Defense.
Trent received his PhD degree in Electrical Engineering from ESAT-MICAS, KU Leuven, Belgium, in 2008, with Dr. Georges Gielen as promotor. His thesis won the international EDAA Outstanding Dissertation Award (#1 in the field for the world). He received a Bachelor’s in Engineering (with great distinction), and a Bachelor’s in Computer Science (with great distinction), both from the University of Saskatchewan, Canada, in 1999. The U of S awarded him the 2001 Outstanding Young Alumni Award “for significant accomplishments since graduation”.
He is the author of the book "Variation-Aware Analog Structural Synthesis: A Computational Intelligence Approach" (Springer, 2009); and the book "Variation-Aware Design of Custom Integrated Circuits: A Hands-on Field Guide" (Springer, 2012). He is co-editor of the books "Genetic Programming Theory and Practice VII" (Springer, 2009) and "Genetic Programming Theory and Practice VIII" (Springer, 2010). He has authored approximately 35 journal papers, book chapters, and conference papers, and has about 20 patents granted / pending. He has been invited to speak at CICC, DAC, ICCAD, MIT, Berkeley, JPL, countless semiconductor companies, and more.
He co-organized the 2009 and 2010 "Genetic Programming Theory and Practice" workshops in Ann Arbor, Michigan. He co-organized the 2012 "International Workshop on Design Automation for Analog and Mixed-Signal Circuits" at ICCAD, San Jose, CA. He serves on the editorial board of the Journal of Genetic Programming and Evolvable Machines, and the Technical Program Committee of CICC (Simulation and Modeling subcommittee). He regularly serves as a reviewer for IEEE Trans. CAD, IEEE Trans. Evolutionary Computation, DAC, DATE, ICCAD, GECCO, CEC, ICES, ACM TODAES, Electronics Letters, and more.