Jiwon Lee

About Me


Profile photo

📍 Suwon, South Korea

I completed my Ph.D. at Yonsei University under the supervision of Professor Won Woo Ro, with a dissertation titled 'High-Performance Address Translation Mechanisms for Graphics Processing Units.' My research interests span a broad range of topics within computer architecture and systems. During my graduate studies, my primary research focus was on designing high-performance address translation methods for GPUs. In addition, I co-authored several research papers on topics such as, enhancing hashed page table, designing efficient CPU microarchitectures, improving memory management methods for unified virtual memory in GPUs, multi-GPU systems, and memory expansion modules. Currently, I am open to postdoctoral opportunities, particularly those that offer the chance to broaden my research expertise beyond my previous focus.

Professional Experience


Staff Engineer Sept. 2024 – Present
Samsung Electronics, Memory Division, South Korea
Graduate Research Assistant Mar. 2018 – Aug. 2024
Yonsei University, Seoul, South Korea
Embedded Systems and Computer Architecture Lab (eSCaL)
Advisor: Professor Won Woo Ro

Education


Ph.D. in Electrical and Electronic Engineering Mar. 2018 – Aug. 2024
Yonsei University, Seoul, South Korea
Dissertation: "High-Performance Address Translation Mechanisms for Graphics Processing Units."
Bachelor of Science in Electrical and Electronic Engineering Mar. 2014 – Feb. 2018
Yonsei University, Seoul, South Korea

Publications


Understanding Distributed Training of Large Language Models with Unified Virtual Memory
Jiwon Lee, and Myung Kuk Yoon
IEEE International Symposium on Workload Characterization (IISWC), 2025. (to appear)
LATPC: Accelerating GPU Address Translation Using Locality-Aware TLB Prefetching and MSHR Compression
Jiwon Lee, Joonsung Kim, Won Woo Ro, and Youngsok Kim
IEEE/ACM International Symposium on Microarchitecture (MICRO), 2025. (to appear)
Heliostat: Harnessing Ray Tracing Accelerators for Page Table Walks [DOI]
Jiwon Lee, Won Woo Ro, and Hyeran Jeon
IEEE/ACM International Symposium on Computer Architecture (ISCA), 2025.
COSMOS: An LLC Contention Slowdown Model for Heterogeneous Multi-core Systems [DOI]
Jiwon Lee, Hyunwuk Lee, and Won Woo Ro
IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2025.
HashScape: Leveraging Virtual Address Dynamics for Efficient Hashed Page Tables [DOI]
Jiwon Lee, Jaewon Kwon, Minjae Kim, and Won Woo Ro
IEEE Transactions on Computers (TC), 2025.
Marching Page Walks: Batching and Concurrent Page Table Walks for Enhancing GPU Throughput [DOI]
Jiwon Lee*, Gun Ko, Myung Kuk Yoon, Ipoom Jeong, Yunho Oh, and Won Woo Ro
IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2025.
REC: Enhancing Fine-Grained Cache Coherence Protocol in Multi-GPU Systems [DOI]
Jiwon Lee, Hongju Kal, Hyunwuk Lee, and Won Woo Ro
Journal of Systems Architecture (JSA), Vol. 160, 103339, Mar. 2025.
Geneva: A Dynamic Confluence of Speculative Execution and In-Order Commitment Windows [DOI]
Jiwon Lee, Jaewon Kwon, Yongju Lee, and Won Woo Ro
IEEE/ACM Design Automation Conference (DAC), 2024.
SnakeByte: A TLB Design with Adaptive and Recursive Page Merging in GPUs [DOI]
Jiwon Lee*, Ju Min Lee, Yunho Oh, William J. Song, and Won Woo Ro
IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2023.
Early-Adaptor: An Adaptive Framework for Proactive UVM Memory Management [DOI]
Jiwon Lee, Myung Kuk Yoon, and Won Woo Ro
IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2023.
Analysis on Memory Access Patterns of Server-Class Workloads in Page- and Cache Line- Granularity
Jiwon Lee, and Won Woo Ro
International Conference on Electronics, Information, and Communication (ICEIC), 2023.
Performance Analysis of Criticality-Aware Out-of-Order Cores for Exploiting MLP [DOI]
Jiwon Lee, and Won Woo Ro
International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), 2023.
Adaptive Data Prefetcher with Probability Learning in LLC [DOI]
Jiwon Lee, and Won Woo Ro
International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), 2023.
Reconstructing Out-of-Order Issue Queue [DOI]
Jiwon Lee, Myung Kuk Yoon, and Won Woo Ro
IEEE/ACM International Symposium on Microarchitecture (MICRO), 2022.
Analysis of DRAM-based Network of DRAM Swap Space Adopting Latency Hiding Technique [DOI]
Jiwon Lee, Jeonghoon Choi, and Won Woo Ro
International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), 2022.
PIMCaffe: Functional Evaluation of a Machine Learning Framework for In-Memory Neural Processing Unit [DOI]
Jiwon Lee, Dongseok Kang, Hongju Kal, and Won Woo Ro
IEEE Access, Vol. 9, pp. 96629–96640, Jul. 2021.
Chapter Six – Deep learning with GPUs [DOI]
Jiwon Lee, Hyunwuk Lee, Dongho Ha, and Won Woo Ro
Advances in Computers, Vol. 122, pp. 167–215, 2021.

Patents


Memory management unit and method of walking page table [Link: US, CN, KR]
Jiwon Lee, Ipoom Jeong, Hongju Kal, Gun Ko, and Hyunwuk Lee
US 18/502,058, CN 202311700848.0, KR 10-2022-0175909
Swap memory device providing data and data block, method of operating the same, and method of operating electronic device including the same [Link: US, CN, KR]
Jiwon Lee, and Jeonghoon Choi
US 18/350,148, CN 202311408257.6, KR 10-2022-0145399
Controller, computing system including the same, and method of creating and searching page table entry for the same [Link: US, CN, KR]
Jiwon Lee
US 17/526,391, CN 202210082121.X, KR 10-2021-0054195
Computing device and method of operating the same
Jiwon Lee
KR 10-2023-0193178
Memory management apparatus and method for UVM
Jiwon Lee
KR 10-2023-0076606
Neural network processing method
Jiwon Lee
KR 10-2022-0041848

List of Ongoing Research


Efficient Memory Management of Large-Scale Memory Space (Under Review)
Exploring System Architectures Using CXL Memory Fabrics (Under Review)
Accelerating Sparse Matrix Computations
Application of IOMMU Specifications
Identifying Bottlenecks in Serverless Environments
Analysis on Different Sparse Matrix Formats

Projects


Development of High Performance Multi-GPU Memory System, National Research Foundation of Korea (NRF), Mar. 2021 Feb. 2024
Development and Application of DRAM-based Memory Hierarchies, Samsung Electronics, Sept. 2020 Aug. 2023
Programmable Neural Network Processors, Samsung Research Funding and Incubation Center for Future Technology, June 2018 May 2022
PIM Core Technology and System Development for Data-intensive Applications, Samsung Electronics, Mar. 2018 Mar. 2020
Development of Reconfigurable Artificial Neural Network Accelerator and Instruction Set Architecture, Korea Electronics Technology Institute (KETI), Mar. 2018 Feb. 2019

Academic Activities


Academic Service
  • External Reviewer Committee, ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Summer Cycle 2026
  • Reviewer, ACM Transactions on Architecture and Code Optimization (TACO), 2025
Mentorship
  • Embedded Systems and Computer Architecture Lab, Yonsei University
    Industry–Academic Mentoring Program, Nov. 2024 Present
  • Intelligence System and Parallel Computer Architecture Lab, Ewha Womans University
    Industry–Academic Mentoring Program, Jan. 2025 Present
Teaching Assistant
  • EEE6611, System Design and Applications Lab, Yonsei University, Fall 2019
  • EEE4473, Embedded System Lab, Yonsei University, Spring 2019
  • EEE3535, Operating Systems, Yonsei University, Fall 2018

Scholarships and Awards


Memory Division Graduate Fellowship, Samsung Electronics, Sept. 2023 Aug. 2024
Excellent Graduate Researcher Scholarship, Yonsei University, Feb. 2023
National Scholarship for Science and Engineering, Korea Student Aid Foundation (KOSAF), Mar. 2014 Feb. 2018
Academic Excellence Award, Yonsei University, Spring 2014

After Hours


Outside of research, I enjoy playing the electric guitar and listening to music rooted in band-style sound. I often rehearse and play cover songs with old bandmates from university, and we've had some great times practicing together in band practice rooms. When I'm alone, I usually pick a song to practice on my own. Being a big fan of rock and heavy metal, my choices tend to be technically demanding and often take months to master. But, when I finally get through them, it's always a very rewarding experience. On top of that, I've also been exploring a broader range of music — from Korean indie bands like Life and Time and various genres, such as fusion and blues artists — and currently, I'm working on Lucretia by Megadeth. This variety helps me stay inspired and continuously expand my musical vocabulary.

Photo of me playing an electric guitar on a stage
Me playing on stage
in Sept. 2018
Photo of my 12-year old electric guitar and amp
My 12-year-old
guitar and amp