Hongsu Byun

Research Interests

• Key-Value Store, Database, Storage System, Data Infrastructure

• Near-Data Processing, Computational Storage, Host-Device Co-Design

• Filesystems, Linux Kernel, Operating Systems, I/O Stacks

• KV-Cache Offloading, AI Serving Systems, LLM Inference

• Retrieval-Augmented Generation (RAG), Vector Databases

Education

• Ph.D. in Computer Science and Engineering, Sogang University

  • Mar. 2021 – Feb. 2026, Seoul, Korea

  • Advisors: Prof. Sungyong Park and Prof. Youngjae Kim

  • Dissertation: Design of High-Performance LSM-tree based Key-Value Store [online]

• B.S. in Computer Science and Engineering, Sogang University

  • Mar. 2015 – Feb. 2021, Seoul, Korea
    (Including Mandatory Military Service, Republic of Korea Army, Dec. 2016 – Sep. 2018)

• Jeju Science High School

  • Mar. 2012 – Feb. 2015, Jeju, Korea

Work Experience

• Mar. 2026 – Present, Postdoctoral Researcher, Sogang University, Seoul, Korea

• Jul. 2022 – Aug. 2022, Visiting Student, Oak Ridge National Laboratory, TN, USA

• May. 2020 – Feb. 2026, Research Assistant, Sogang University, Seoul, Korea

Selected Publications (Full Publication List) 

*Co-first author, †Co-corresponding author, KIISE S (한국정보과학회 최우수), KIISE A (한국정보과학회 우수)

• Dual-Blade: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference,
  Bodon Jeong, Hongsu Byun, Youngjae Kim, Weikuan Yu, Kyungkeun Lee, Jihoon Yang, Sungyong Park,
  In Proceedings of IEEE International Conference on Distributed Computing Systems (ICDCS) (2026), Seoul, South Korea, June 22-25, 2026. (Acceptance rate: 18.59%) [KIISE A]

• QUICKTOPIA: Iteration-Level GPU Frequency Control for Energy–Latency Co-Optimization in LLM Inference,
  Soyang Baek, Bodon Jeong, Hongsu Byun, Sungyong Park,
  In Proceedings of the IEEE International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2026), Sydney, Australia, May 18-21, 2026. (Acceptance Rate: 62/247 = 25.1%) [KIISE A]

• [KIISE S] Resystance: Unleashing Hidden Performance of Compaction in LSM-trees via eBPF,
  Hongsu Byun*, Seungjae Lee*, Honghyeon Yoo, Myoungjoon Kim, Sungyong Park,
  In Proceedings of the IEEE International Conference on Data Engineering (ICDE) (2026), Montréal, Canada, May 4-8, 2026. [arXiv][news]

• PACMAN: Power Usage-Aware Capping  and Management for All-Flash Storage Servers,
  Bodon Jeong, Hongsu Byun, Kyungkeun Lee, Bumjun Kim, Jeong-Uk Kang, Sungyong Park, 
  In Proceedings of the International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS) (2025), Paris, France, October 21-23, 2025. (Acceptance rate: 24%) [KIISE A]

• Revisiting Multi-threaded Compaction in LSM-trees: Enabling Compaction Pipelining,
  Hongsu Byun*, Honghyeon Yoo*, Sungyong Park,
  In Proceedings of the International Conference on Parallel Processing (ICPP) (2025), San Diego, CA, USA, September 8-11, 2025.
  (Acceptance rate: 78/292 = 26.7%) [KIISE A] [online][news]

• KVAccel: A Novel Write Accelerator for LSM-Tree-Based KV Stores with Host-SSD Collaboration,
  KiHwan Kim, Hyunsun Chung, Seonghoon Ahn, Junhyeok Park, Safdar Jamil, Hongsu Byun, Myungcheol Lee, Jinchun Choi, Youngjae Kim,
  In Proceedings of the IEEE International Parallel and Distributed Processing Symposium (IPDPS) (2025), Milan, Italy, June 3-7, 2025 [KIISE A]

• ECO-KVS: Energy-aware Compaction Offloading Mechanism for LSM-tree based Key-Value Stores in Edge Federation,
  Jeeseob Kim*, Hongsu Byun*, Seungjae Lee, Myoungjoon Kim, Youngjae Kim, Zaipeng Xie, Sungyong Park,
  In Proceedings of the IEEE International Symposium on Cluster, Cloud and Grid Computing (CCGRID) (2025), Tromsø, Norway, May 19-22, 2025. (Sustainable IT Systems Track Acceptance rate: 6/31 = 19.3%, Overall Acceptance Rate: 55/219 = 25.1%) [KIISE A] [online][pdf] [talk] [news]

• Refining Compaction Offloading I/O Stack for LSM-based Key-value Stores with SPDK,
  Honghyeon Yoo*, Hongsu Byun*, Sungyong Park,
  In Proceedings of the IEEE International Conference on Cluster Computing (CLUSTER) (Poster) (2024), September 24-27, 2024. [online]

• An Analytical Model-based Capacity Planning Approach for Building CSD-based Storage Systems,
  Hongsu Byun, Safdar Jamil, Jungwook Han, Sungyong Park, Myungcheol Lee, Changsoo Kim, Beongjun Choi, Youngjae Kim,
  ACM Transactions on Embedded Computing Systems (TECS) (2024) Vol.23, Issue 6, No. 92, pp. 1-25, September 11, 2024. [online][pdf]

• Coordinating Compaction between LSM-tree based Key-Value Stores for Edge Federation,
  Jeeseob Kim, Honghyeon Yoo, Seungjae Lee, Hongsu Byun†, Sungyong Park†,
  In Proceedings of the IEEE International Conference on Cloud Computing (CLOUD) (2024), Shenzhen, China, July 7-13, 2024. (Acceptance rate: 19.4%) [KIISE A][online][news]

• Scalable Filesystem : High-performance Linux Filesystem and Open-source Kafka-based Streaming Big Data Processing Service [video]