Opportunities and challenges in accelerating large-scale search using NVM-based in-memory computing

Speaker: Xiaobo Sharon Hu, University of Notre Dame

Abstract: Large-scale search plays a crucial role in a broad range of machine learning applications, where achieving fast and energy-efficient search is highly desirable. However, accelerating search at scale remains challenging due to massive data volumes and irregular memory access patterns. In-memory computing architectures—particularly non-volatile memory-based content addressable memory (CAM)—offers a promising solution by enabling massively parallel search with reduced energy consumption. This talk examines how architectural and system-level design choices impact the overall efficiency, scalability, and performance of CAM-based search accelerators. I will highlight bottlenecks and key tradeoffs in such accelerators and present several algorithm-hardware codesign techniques that improve performance and energy efficiency. The discussions will offer insights for hardware designers and system architects working toward next-generation high-performance accelerators for large-scale search and retrieval workloads.

Biography: Xiaobo Sharon Hu is Leo E. and Patti Ruth Linbeck Professor of Engineering in the department of Computer Science and Engineering at the University of Notre Dame, USA. Her research interests include low-power and reliable system design, circuit and architecture design with emerging technologies, real-time embedded systems, and hardware-software co-design. She has published more than 500 papers in these areas and received best paper awards from top design automation conferences. She served as the General Chair and/or TPC Chair of Design Automation Conference, Real-Time Systems Symposium, Embedded Systems Week, etc. She was the Editor-in-Chief of ACM Transactions on Design Automation of Electronic Systems and served as Associate Editor of other ACM and IEEE journals.  Sharon Hu is a Fellow of the ACM and the IEEE.

Host: Prof. Laura Pozzi