Hyunkyun Shin
M.S. Student

Research Interests

  • GPU Unified Virtual Memory
  • Memory system, architecture

Education

  • 2025.02~, Combined M.S. and Ph.D. Program, Dept. of Electrical and Electronic Engineering, Yonsei University
  • 2024.02~2025.02, M.S., Student, Dept. of Electrical Engineering and Computer Science, DGIST
  • 2020.02~2024.02, Bachelor of Convergence engineering(Computer Science and Engineering Major Track, Mechanical Engineering Minor Track), School of Basic Science, DGIST

Publications

* Corresponding Author
MICRO | TOP-Tier | GPU Next-Gen Memory System
ReclaimX: Device-Side Memory Reclamation via Stalled GPU Execution for UVM Oversubscription

Seongtae Bang, Hyunkyun Shin, Hyungwon Park, Minho Kim, and Daehoon Kim*

IEEE/ACM International Symposium on Microarchitecture (MICRO), 2026

MICRO | TOP-Tier | AI System
AMG: AMX–GPU Cooperative Acceleration of Billion-Scale ANNS via GEMM Reformulation

Minho Kim, Houxiang Ji, Hwanjun Lee, Jaeyoung Kang, Sungju Kim, Hyunkyun Shin, Daehoon Kim*, and Nam Sung Kim

IEEE/ACM International Symposium on Microarchitecture (MICRO), 2026

HPCA | TOP-Tier | GPU Next-Gen Memory System
ARIADNE: Adaptive UVM Management for Efficient GPU Memory Oversubscription

Hyunkyun Shin, Seongtae Bang, Hyungwon Park, and Daehoon Kim*

IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2026

CAL | SCI | GPU Next-Gen Memory System
SAFE: Sharing-aware Prefetching for Efficient GPU Memory Management with Unified Virtual Memory

Hyunkyun Shin, Seongtae Bang[co-first author], Hyungwon Park, and Daehoon Kim*

IEEE Computer Architecture Letters (CAL), 2025