Abstract: The memory system is a major bottleneck in modern computing systems. We are experiencing the large negative performance, energy and sustainability impacts of the memory bottleneck especially today with the continued growth of large-scale machine learning workloads that demand increasingly more data. At the same time, we are also experiencing growing difficulties in designing the increasingly complex memory system (and the storage system) to achieve high performance and efficiency. This talk covers two major promising directions in computer design that are synergistic with each other: the use of ML to design better memory systems and the customization and rethinking of the memory system to better execute ML (as well as other data-intensive) workloads. In the first direction, we will cover some recent works where we use machine learning to perform better decision making in controllers used in memory systems, including ML-driven (e.g., reinforcement learning based) intelligent memory controllers, prefetchers, cache miss predictors, and storage management systems. We will show that using online machine learning to drive decisions made by such controllers can lead to large performance and efficiency benefits for a wide variety of workloads. In the second direction, if time permits, we will provide an overview of our efforts in memory-centric computing and acceleration of major workloads, including generative artificial intelligence, machine learning, graph analytics, genomics, databases.

Bio: Onur Mutlu is a Professor of Computer Science at ETH Zurich. He previously held the William D. and Nancy W. Strecker Early Career Professorship at Carnegie Mellon University. His current research interests are in computer architecture, computing systems, hardware security, memory & storage systems, and bioinformatics, with a major focus on designing fundamentally energy-efficient, high-performance, and robust computing systems. Many techniques he, with his group and collaborators, has invented over the years have largely influenced industry and have been widely employed in commercial microprocessors and memory & storage systems used daily by hundreds of millions of people. He obtained his PhD and MS in ECE from the University of Texas at Austin and BS degrees in Computer Engineering and Psychology from the University of Michigan, Ann Arbor. He started the Computer Architecture Group at Microsoft Research (2006-2009), and held product, research and visiting positions at Intel Corporation, Advanced Micro Devices, VMware, Google, and Stanford University. He received various honors for his research, including recently the 2025 IEEE Computer Society Harry H. Goode Memorial Award “for seminal contributions to computer architecture research and practice, especially in memory systems,” 2024 IFIP WG10.4 Jean-Claude Laprie Award in Dependable Computing (for the original RowHammer work), 2022 Persistent Impact Prize of the Non-Volatile Memory Systems Workshop (for original architectural work on Phase Change Memory), 2021 IEEE High Performance Computer Architecture Conference Test of Time Award (for the Runahead Execution work) and tens of best paper or “Top Pick” paper recognitions at various leading computer systems, architecture, and security venues. He is an ACM Fellow, IEEE Fellow, and an elected member of the Academy of Europe. He enjoys teaching, mentoring, and enabling & democratizing access to high-quality research and education. He has supervised 24 PhD graduates, multiple of whom received major dissertation awards, 15 postdoctoral trainees, and more than 60 Master’s and Bachelor’s students. His computer architecture and digital logic design course lectures and materials are freely available on YouTube (https://www.youtube.com/OnurMutluLectures & https://www.youtube.com/@CMUCompArch), and his research group (https://safari.ethz.ch/) makes a wide variety of open-source artifacts freely available online (https://github.com/CMU-SAFARI). For more information, please see his webpage at https://people.inf.ethz.ch/omutlu/.