Jeongseob Ahn (안정섭)

Associate Professor @ Korea University

School of Electrical Engineering (Computer Division)

Research Group: Computer Systems Laboratory

E-mail: jsahn [AT] csl.korea.ac.kr

Tel: +82-2-3290-3229

Office: New Engineering Building #518

Short Bio

Jeongseob Ahn is an associate professor in the School of Electrical Engineering (Computer Division) at Korea University. He is also affiliated with the Department of Communications Engineering. His research interests lie in building efficient computer systems. Before joining Korea University, he served as a faculty member at Ajou University for six and a half years. Prior to that, he worked at Oracle Labs in California, where he contributed to the development of a large-scale data analytics system with specialized hardware, known as RAPID. He received his PhD in Computer Science from KAIST in 2015, under the supervision of Prof. Jaehyuk Huh, and earned his BS in Computer Science and Engineering from Dongguk University in 2009. During his undergraduate studies, he was a member of Samsung Software Membership.

Research Interests

I am actively looking for highly self-motivated students interested in building fast and efficient computer systems for AI, cloud, and emerging hardware. Please reach out to me with your CV and transcript.

Teaching (Spring 2025)

Recent Publications (full list)

Accelerating LLM Serving for Multi-turn Dialogues with Efficient Resource Management [ Paper | Slides ]

Jinwoo Jeong and Jeongseob Ahn

ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), March-April 2025

GPU-centric Memory Tiering for LLM Serving with NVIDIA Grace Hopper Superchip [ Paper ]

Woohyung Choi, Jinwoo Jeong, Hanhwi Jang, and Jeongseob Ahn

IEEE Computer Architecture Letters (CAL), 24(1), January-June 2025

Accelerating Page Migrations in Operating Systems with Intel DSA [ Paper ]

Jongho Baik, Jonghyeon Kim, Chang Hyun Park, and Jeongseob Ahn

IEEE Computer Architecture Letters (CAL), 24(1), January-June 2025

EnvPipe: Performance-preserving DNN Training Framework for Saving Energy [ Paper | Slides | Video | Code ]

Sangjin Choi, Inhoe Koo, Jeongseob Ahn, Myeongjae Jeon, and Youngjin Kwon

USENIX Annual Technical Conference (ATC), July 2023

Fast and Efficient Model Serving Using Multi-GPUs with Direct-Host-Access [ Paper | Slides | Code ]

Jinwoo Jeong, Seungsu Baek, and Jeongseob Ahn

ACM European Conference on Computer Systems (EuroSys), May 2023

Gilles Muller Best Artifact Award

Academic Services

Program Committee

Organizing Committee