ECE Distinguished Speaker Series: Suman Datta, “The Changing Paradigm of Semiconductor Research”

The rapid proliferation and diffusion of large language model driven agentic inference continue to fuel an unprecedented demand for semiconductors for high-performance compute. Addressing this compute demand goes beyond just CMOS logic transistor scaling and requires a system-driven rethink of semiconductor research. In this talk, we explore emerging semiconductor technology pillars essential for the next gen computing. In the first pillar, we will discuss Complementary FETs (CFET) utilizing next-generation gate stacks processed under low thermal budgets. In the next pillar, we will address the impending SRAM Scaling Wall by transitioning to fast memory alternatives that surpass the density limitations of traditional SRAM. In the third pillar, we will present a thermally resilient persistent memory based on ferroelectric materials to implement 3D memory-on-logic hardware. In the final pillar, we will present a new hardware that integrates logic functions within memory arrays to support low latency iterative AI reasoning. We will end with a discussion on how the introduction of these technologies necessitates a deeper understanding of reliability physics and characteristics.
Bio: Suman Datta received the B.Tech degree in Electrical Engineering from the Indian Institute of Technology, Kanpur, India and the Ph.D degree in Electrical and Computer Engineering from University of Cincinnati, Cincinnati, Ohio. He joined Georgia Tech’s School of Electrical and Computer Engineering (ECE) as the Joseph M. Pettit Chair and Georgia Research Alliance (GRA) Eminent Scholar in the Fall of 2022. From 2016 till 2022, he was the Stinson Endowed Chair Professor of Nanotechnology in the Department of Electrical Engineering at the University of Notre Dame, Notre Dame, Indiana, where he directed research in the Nanoelectronic Devices and Circuits Laboratory. He was the Director of the multi-university $40.4M SRC and DARPA sponsored Applications and Systems-driven Center for Energy-efficient integrated NanoTechnologies (ASCENT) till Dec 2022. In addition, he was the Director of the SRC and NSF sponsored Center for Extremely Energy Efficient Collective Electronics (EXCEL). His research involves heterogenous computing, brain-inspired computing, in-memory/near-memory computing and cryogenic computing using advanced CMOS and beyond CMOS devices.