Research Experience

Computer Architecture and System Software Lab
(ComSys)

Korea University
Mar. 2024 - Feb. 2028
Ph.D Student | Advisor: Yunho Oh

Research Focus

  • Energy-efficient DNN accelerator architecture
  • Software techniques for improving the performance of GPUs
  • Analysis on CXL-based memory systems for datacenter servers

Computer Architecture and System Software Lab
(ComSys)

SungKyunKwan University
Mar. 2022 - Feb. 2024
M.S. Student | Advisor: Yunho Oh

Research Focus

  • Energy-efficient memory systems for DNN accelerator architecture
  • DNN accelerator architecture supporting arbitrary numeric formats
Thesis: Improving energy efficiency of DNN accelerators with a new memory hierarchy

Publications

(C=Conference, J=Journal)

[J.3] Jaeyong Park, Sangun Choi, Jongmin Kim, Gunjae Koo, Myung Kuk Yoon, and Yunho Oh. (2025). TM-Training: An Energy-Efficient Tiered Memory System for Deep Learning Training in NPUs. ACM Transactions on Storage.
[J.2] Minseong Gil, Jaebeom Jeon, Junsu Kim, Sangun Choi, Gunjae Koo, Seokin Hong, Myung Kuk Yoon, and Yunho Oh. (2024). TLP Balancer: Predictive Thread Allocation for Multi-Tenant Inference in Embedded GPUs. IEEE Embedded Systems Letters.
[J.1] Sangun Choi, Seongjun Park, Jaeyong Park, Jongmin Kim, Gunjae Koo, Seokin Hong, Myung Kuk Yoon, and Yunho Oh. (2024). SAVector: Vectored Systolic Arrays. IEEE Access.

Education

SungKyunKwan University (SKKU)

Mar. 2022 - Feb. 2024

M.S. degree in electrical and computer engineering

Advisor: Yunho Oh

GPA: 4.44/4.5 (3.95/4.00)

Korea National University of Transportation (KNUT)

Mar. 2018 - Aug. 2021

B.S. degree in railroad electrical and electronics engineering

GPA: 4.39/4.5 (3.90/4.00)

Skills

Languages: Korean (Native), English (Professional)

Programming Languages: C, C++, Python, Shell script

Software Tools: SCALE-Simv2, Accelergy, CACTI7.0, GPGPU-Sim

Professional Services

Artifact Evaluation Committee

  • 30th International Symposium on High-Performance Computer Architecture (HPCA), 2024