Research Interests

• Computer Architecture and Systems for Accelerating Machine Learning and Emerging Applications

Education

• Doctor of Philosophy, Computer Science, Stanford University, Sep. 2024 - Present

  • Advisor: Kunle Olukotun 

• Master of Science, School of Computing, KAIST, Feb. 2023

  • Thesis: Improving the Utilization of NPU On-chip Memory with Computation Rearrangement for DNN Training

  • Advisor: Jaehyuk Huh 

• Bachelor of Science, School of Computing, KAIST, Feb. 2021

  • Served in the military for two years

  • Minor Degree in Mathematical Sciences

Employment

• Research Scientist, School of Computing, KAIST, Mar. 2023 - Aug. 2024

  • Advisor: Jaehyuk Huh 

Publications

• Seonjin Na, Jungwoo Kim, Sunho Lee, and Jaehyuk Huh, “Supporting Secure Multi-GPU Computing with Dynamic and Batched Metadata Management”, the 30th IEEE International Symposium on High-Performance Computer Architecture (HPCA), March 2024 [paper][slide]

• Jaemin Hong, Sunghwan Shim, Sanguk Park, Taewoo Kim, Jungwoo Kim, Junsoo Lee, Sukyoung Ryu, and Jeehoon Kang, “Taming shared mutable states of operating systems in Rust”, Science of Computer Programming (SCICO), 2024 [paper][artifact]

• Jungwoo Kim, Seonjin Na, Sanghyeon Lee, Sunho Lee, and Jaehyuk Huh, “Improving Data Reuse in NPU On-chip Memory with Interleaved Gradient Order for DNN Training”, the 56th IEEE/ACM International Symposium on Microarchitecture (MICRO), October 2023 [paper][slide]

• Soojin Hwang*, Sunho Lee*, Jungwoo Kim, Hongbeen Kim, and Jaehyuk Huh, "mNPUsim: Evaluating the Effect of Sharing Resources with Multi-Core NPUs", 2023 IEEE International Symposium on Workload Characterization (IISWC), October 2023 (*: equal contribution) [paper][slide][github]

• Sunho Lee, Seonjin Na, Jungwoo Kim, Jongse Park, and Jaehyuk Huh, "Tunable Memory Protection for Secure Neural Processing Units", the 40th IEEE International Conference on Computer Design (ICCD), October, 2022 [paper][slide]

• Sunho Lee, Jungwoo Kim, Seonjin Na, Jongse Park, and Jaehyuk Huh, "TNPU: Supporting Trusted Execution with Tree-less Integrity Protection for Neural Processing Unit", the 28th IEEE International Symposium on High-Performance Computer Architecture (HPCA), April, 2022 [paper][slide]

Research Experiences

• Stanford Pervasive Parallelism Lab, California, USA, Sep. 2024 - Present (Advisor : Kunle Olukotun)

  • Dynamism in Machine Learning Accelerators

• KAIST CASYS Lab, Daejeon, Republic of Korea, Jan. 2021 - Aug. 2024 (Advisor : Jaehyuk Huh)

  • Accelerator Performance

    • Operation Scheduling for High-performance DNN training in NPU (MICRO 2023)

    • mNPUsim: Open-sourced Cycle-accurate Multi-core NPU Simulator (IISWC 2023)

  • Accelerator Hardware-based Security

    • Efficient Communication Mechanism for Secure Multi-GPU Architecture (HPCA 2024)

    • Trusted Execution Environment Architecture for NPU (HPCA 2022)

    • Optimized Hardware Memory Protection for secure NPU (ICCD 2022)

• KAIST CP Lab, Daejeon, Republic of Korea, Jun. 2020 - Jan. 2021 (Advisor : Jeehoon Kang) 

  • Reimplementation of xv6 (operating system written in C) using Rust (SCICO 2024) : rv6

Patents

• [Application: KR 1020230055347] Jaehyuk Huh, Jungwoo Kim, Seonjin Na, Sanghyeon Lee, and Sunho Lee, "Improving the Utilization of NPU On-chip Memory with Computation Rearrangement for DNN Training", Korean Patent.

• [Application: KR 1020230055346] Jaehyuk Huh, Seonjin Na, Jungwoo Kim, and Sunho Lee, "Dynamic One-time Pad Table Management for Secure Multi-GPU Communication", Korean Patent.

Awards & Scholarships

• National Full Graduate Scholarship, KAIST, 2021 - 2023

• Third Place, Best Paper Awards, KAIST-Samsung Electronics Industry-Academic Exchange, 2022

• Excellent Teaching Assistant Award, School of Computing, KAIST, 2022

• Cum Laude, College of Engineering, KAIST, 2021 

• Dean's List, College of Engineering, KAIST, Fall 2018 & Spring 2019

• National Science & Technology Scholarship, Korea Student Aid Foundation, 2019

• LINE Scholarship, LINE corporation, 2019

• National Full Undergraduate Scholarship, Korea Student Aid Foundation, 2014 - 2020

Teaching Experiences

• School of Computing, KAIST

  • Teaching Assistant for Computer Organization (CS311) : Spring 2022

  • Teaching Assistant for System Programming (CS230) : Fall 2021

  • Teaching Assistant for Introduction to Programming (CS101) : Fall 2020, Spring 2021

• School of Freshman, KAIST

  • Tutor for Calculus II (MAS102) : Fall 2015, Spring 2016

• Global Leadership Center, KAIST

  • Instructor for Humanity/Leadership III<Your Story: Make Your Own Lecture> (HSS062): Fall 2015, Spring 2016