Profile

Myeonggyun Han

Myeonggyun Han

Assistant Professor @ Kyungpook National University (KNU)
School of Computer Science and Engineering
Research Group: Computer Systems and AI Lab (CoSAIL)
E-mail: mhan [AT] knu.ac.kr
Office: Rm. 413-1, Bldg. 414

Short Bio

Myeonggyun Han is an assistant professor in the School of Computer Science and Engineering at Kyungpook National University (KNU), where he leads the Computer Systems and AI Lab (CoSAIL). His research interests lie in computer systems and system software for AI, with a focus on efficient large language model (LLM) inference, on-device and edge AI systems, and system software support for emerging AI workloads. He received B.S. and Ph.D. degrees in Computer Science and Engineering from Ulsan National Institute of Science and Technology (UNIST), under the supervision of Prof. Woongki Baek. Prior to joining KNU, he worked as a researcher at the Electronics and Telecommunications Research Institute (ETRI).

Research Interests

  • Computer Systems and System Software for AI
  • Efficient LLM Inference and Runtime Optimization
  • Resource Management for On-Device and Edge AI Systems

We are looking for highly motivated students interested in computer systems and AI systems research. Students with interests in operating systems, computer architecture, machine learning systems, and system software for AI are encouraged to contact me with your CV and transcript.

News

  • May. 2026: Our paper entitled “MemSpec: Memory-Aware Runtime for Adaptive Draft Scheduling in Speculative Decoding on Edge Devices” has been accepted to LCTES 2026.

  • Dec. 2025: Our lab has been selected for the Advanced GPU Utilization Support Program Beta Service and has been granted access to sixteen NVIDIA B200 GPUs.

  • Aug. 2025: Prof. Han has been awarded the Outstanding Young Scientist Grants by the National Research Foundation of Korea (NRF). Our project, “Input Complexity-Aware Adaptive Execution and Memory Management for Efficient On-Device LLM Inference”, will be funded from September 2025 to August 2026.

  • Jun. 2025: We’re excited to share that our lab has been selected for the Google Research Credits Program (Jun 2025 – Jun 2026).

Publications

  • Eunjeong Kim, Yeong Jun Jeon, and Myeonggyun Han, “MemSpec: Memory-Aware Runtime for Adaptive Draft Scheduling in Speculative Decoding on Edge Devices,” in the Proceedings of the 27th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES), Jun. 2026 (to appear).

  • Myeonggyun Han, Eunseong Park, Youngsam Shin, Deok-Jae Oh, YeonGon Cho, and Woongki Baek, “COSMOS: Coordinated Management of Cores, Memory, and Compressed Memory Swap for QoS-Aware and Efficient Workload Consolidation for Memory-Intensive Applications,” in the IEEE Access (Access), Nov. 2023.

  • Sowoong Kim, Myeonggyun Han, and Woongki Baek, “MARF: A Memory-Aware CLFLUSH-Based Intra- and Inter-CPU Side-Channel Attack,” in the Proceedings of the 28th European Symposium on Research in Computer Security (ESORICS), Sep. 2023.

  • Myeonggyun Han and Woongki Baek, “SDRP: Safe, Efficient, and SLO-Aware Workload Consolidation through Secure and Dynamic Resource Partitioning,” in the IEEE Transactions on Services Computing (TSC), July-Aug. 2022.

  • Sowoong Kim, Myeonggyun Han, and Woongki Baek, “DPrime+DAbort: A High-Precision and Timer-Free Directory-Based Side-Channel Attack in Non-Inclusive Cache Hierarchies using Intel TSX,” in the Proceedings of the 28th IEEE International Symposium on High-Performance Computer Architecture (HPCA), Apr. 2022.

  • Myeonggyun Han and Woongki Baek, “HERTI: a Reinforcement Learning-Augmented System for Efficient Real-Time Inference on Heterogeneous Embedded Systems,” in the Proceedings of the 30th International Conference on Parallel Architectures and Compilation Techniques (PACT), Sept. 2021.

  • Jinsu Park, Seongbeom Park, Myeonggyun Han, and Woongki Baek, “PALM: Progress- and Locality-Aware Adaptive Task Migration for Efficient Thread Packing,” in the Proceedings of the 35th IEEE International Parallel and Distributed Processing Symposium (IPDPS), May 2021.

  • Myeonggyun Han, Jinsu Park, and Woongki Baek, “Design and Implementation of a Criticality- and Heterogeneity-Aware Runtime System for Task-Parallel Applications,” in the IEEE Transactions on Parallel and Distributed Systems (TPDS), May 2021.

  • Myeonggyun Han, Jihoon Hyun, Seongbeom Park, and Woongki Baek, “Hotness- and Lifetime-Aware Data Placement and Migration for High-Performance Deep Learning on Heterogeneous Memory Systems,” in the IEEE Transactions on Computers (TC), Mar. 2020.

  • Myeonggyun Han, Jihoon Hyun, Seongbeom Park, Jinsu Park, and Woongki Baek, “MOSAIC: Heterogeneity-, Communication-, and Constraint-Aware Model Slicing and Execution for Accurate and Efficient Inference,” in the Proceedings of the 28th International Conference on Parallel Architectures and Compilation Techniques (PACT), Sept. 2019.

  • Jinsu Park, Seongbeom Park, Myeonggyun Han, and Woongki Baek, “POSTER: The Performance Impact of Thread Packing on Synchronization-Intensive Applications,” in the Proceedings of the 28th International Conference on Parallel Architectures and Compilation Techniques (PACT), Sept. 2019 (Poster).

  • Jinsu Park, Seongbeom Park, Myeonggyun Han, Jihoon Hyun, and Woongki Baek, “HyPart: A Hybrid Technique for Practical Memory Bandwidth Partitioning on Commodity Servers,” in the Proceedings of the 27th International Conference on Parallel Architectures and Compilation Techniques (PACT), Nov. 2018.

  • Myeonggyun Han, Seongdae Yu, and Woongki Baek, “Secure and Dynamic Core and Cache Partitioning for Safe and Efficient Server Consolidation,” in the Proceedings of the 18th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing (CCGrid), May. 2018 (selected as a Best Paper Finalist).

  • Myeonggyun Han, Jinsu Park, and Woongki Baek, “CHRT: a Criticality- and Heterogeneity-Aware Runtime System for Task-Parallel Applications,” in the Proceedings of the 2017 Design, Automation & Test in Europe Conference & Exhibition (DATE), Mar. 2017.

  • Jinsu Park, Myeonggyun Han, and Woongki Baek, “Quantifying the Performance Impact of Large Pages on In-Memory Big-Data Workloads,” IEEE International Symposium on Workload Characterization (IISWC), Sept. 2016.

Honors and Awards

  • NRF Research Grant (Outstanding Young Scientist Grants), 2025

  • NAVER Ph.D. Fellowship Award, 2020
    • The fellowship supports graduate students who have published excellent research papers on computing topics (5M KRW scholarship)

  • Student Travel Grant, PACT 2019
    • 28th International Conference on Parallel Architectures and Compilation Techniques (PACT)

  • Best Paper Finalist, CCGrid 2018

  • Silver Prize, NAVER UNIST Undergraduate Poster Award 2017
    • The second place among the 31 participating teams

  • Finalist, 2014 Korea Whitehat Contest (Team HeXA)
    • Held by Ministry of National Defense and National Intelligence Service
    • 5th place at preliminary contest