Exploring High-Bandwidth Flash (HBF) for Modern LLM-Serving Systems
Join us for our upcoming Future Computing Seminar Series
Speaker: Prof. Jisung Park, Pohang University of Science and Technology (POSTECH)
Date: Friday, July 24th, 2026, 14:30 CET
Where: ETZ F91
Abstract:
This talk discusses the potential benefits and technical challenges of an emerging technology called High-Bandwidth Flash (HBF), as a scalable complement to DRAM-based HBM for overcoming the memory-capacity wall in modern LLM-serving. The rapid scaling of LLMs is creating unprecedented demand for memory capacity and bandwidth that conventional HBM alone cannot meet efficiently. NAND flash memory is gaining attention as a promising solution for extending GPU memory at a lower cost. For example, SanDisk first proposed the concept of HBF in 2025, targeting ~20x larger capacity with matching bandwidth compared to DRAM-based HBM, and major memory manufacturers are now actively exploring its realization. In this talk, I will present our ongoing research to systematically explore the potential benefits of using HBF as the main GPU-memory component, as well as the technical challenges that must be addressed to realize those benefits.
Speaker Bio:
Jisung Park is an Assistant Professor in the Department of Computer Science and Engineering at POSTECH, where he co-leads the Computer Architecture and Operating Systems (CAOS) Laboratory. He is also affiliated with the Graduate School of Artificial Intelligence. Prior to joining POSTECH in 2022, he spent three years as a Senior Researcher and Lecturer in the Department of Information Technology and Electrical Engineering at ETH Zürich, working with the SAFARI Research Group led by Prof. Onur Mutlu. He earned his Ph.D. degree in Electrical Engineering and Computer Science from Seoul National University. His research interests lie in a broad range of computer systems, including computer architecture, system software, system security, and HW/SW cooperation, with a special focus on memory and storage systems.