I am a graduate student in the Dept. of Electrical and Computer Engineering (ECE) at Seoul National University. My research interests are systems and architectures for AI, such as distributed AI systems and AI accelerators. I’m also interested in all system- and architecture-related researches :)

Publication

¹ indicates co-first authors.

Piccolo: Large-Scale Graph Processing with Fine-Grained In-Memory Scatter-Gather

Changmin Shin, Jaeyong Song, Hongsun Jang, Dogeun Kim, Jun Sung, Taehee Kwon, Jae Hyung Ju, Frank Liu, Yeonkyu Choi, Jinho Lee

International Symposium on High-Performance Computer Architecture (HPCA) · Mar 2025 · To appear

GraNNDis: Fast Distributed Graph Neural Network Training Framework for Multi-Server Clusters

Jaeyong Song, Hongsun Jang, Hunseong Lim, Jaewon Jung, Youngsok Kim, Jinho Lee

The International Conference on Parallel Architectures and Compilation Techniques (PACT) · Oct 2024 · To appear

A Case for In-Memory Random Scatter-Gather for Fast Graph Processing

Changmin Shin¹, Taehee Kwon¹, Jaeyong Song, Jae Hyung Ju, Frank Liu, Jinho Lee

IEEE Computer Architecture Letters (CAL) · Mar 2024 · 10.1109/LCA.2024.3376680

PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor

Jaewon Jung, Hongsun Jang, Jaeyong Song, Jinho Lee

Computer Vision and Pattern Recognition Conference (CVPR) · Mar 2024 · 10.48550/arXiv.2403.06668

AGAThA: Fast and Efficient GPU Acceleration of Guided Sequence Alignment for Long Read Mapping

Seongyeon Park, Junguk Hong, Jaeyong Song, Hajin Kim, Youngsok Kim, Jinho Lee

ACM Annual Symposium on Principles and Practice of Parallel Programming (PPoPP) · Feb 2024 · 10.1145/3627535.3638474

Pipette: Automatic Fine-Grained Large Language Model Training Configurator for Real-World Clusters

Jinkyu Yim¹, Jaeyong Song¹, Yerim Choi, Jaebeen Lee, Jaewon Jung, Hongsun Jang, Jinho Lee

Design, Automation and Test in Europe Conference (DATE) · Mar 2024 · 10.23919/DATE58400.2024.10546826

Smart-Infinity: Fast Large Language Model Training using Near-Storage Processing on a Real System

Hongsun Jang, Jaeyong Song, Jaewon Jung, Jaeyoung Park, Youngsok Kim, Jinho Lee

Best Paper Award-Honorable Mention, International Symposium on High-Performance Computer Architecture (HPCA) · Mar 2024 · 10.1109/HPCA57654.2024.00034

Pipe-BD: Pipelined Parallel Blockwise Distillation

Hongsun Jang, Jaewon Jung, Jaeyong Song, Joonsang Yu, Youngsok Kim, Jinho Lee

2023 Design, Automation & Test in Europe Conference & Exhibition (DATE) · Apr 2023 · 10.23919/DATE56975.2023.10137044

Optimus-CC: Efficient Large NLP Model Training with 3D Parallelism Aware Communication Compression

Jaeyong Song¹, Jinkyu Yim¹, Jaewon Jung, Hongsun Jang, Hyung-Jin Kim, Youngsok Kim, Jinho Lee

Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2 · Jan 2023 · 10.1145/3575693.3575712

SGCN: Exploiting Compressed-Sparse Features in Deep Graph Convolutional Network Accelerators

Mingi Yoo¹, Jaeyong Song¹, Jounghoo Lee, Namhyung Kim, Youngsok Kim, Jinho Lee

2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA) · Feb 2023 · 10.1109/HPCA56546.2023.10071102

Slice-and-Forge: Making Better Use of Caches for Graph Convolutional Network Accelerators

Mingi Yoo¹, Jaeyong Song¹, Hye Yoon Lee, Jounghoo Lee, Namhyung Kim, Youngsok Kim, Jinho Lee

PACT (Best Paper Award) · Oct 2022 · 10.1145/3559009.3569693

Making a Better Use of Caches for GCN Accelerators with Feature Slicing and Automatic Tile Morphing

Mingi Yoo¹, Jaeyong Song¹, Jounghoo Lee, Namhyung Kim, Youngsok Kim, Jinho Lee

IEEE Computer Architecture Letters · Jul 2021 · 10.1109/LCA.2021.3090954

Fast Adversarial Training with Dynamic Batch-level Attack Control

Jaewon Jung, Jaeyong Song, Hongsun Jang, Hyeyoon Lee, Kanghyun Choi, Noseong Park, Jinho Lee

Design Automation Conference (DAC) · Jul 2023 · 10.1109/DAC56929.2023.10247930

Education

Seoul National University
- Ph.D student in Electrical and Computer Engineering (September 2023 -)
Yonsei University
- M.S. in Computer Science (August 2023)
- B.E. in Applied Statistics (August 2021)
- B.S. in Computer Science (August 2021)
- minor in Education (August 2021)

Awards

Best Paper Award (October 2022)
- Slice-and-Forge: Making Better Use of Caches for Graph Convolutional Network Accelerators (PACT’22)
Best Paper Award Honorable Mention (March 2024)
- Smart-Infinity: Fast Large Language Model Training using Near-Storage Processing on a Real System (HPCA’24)
Samsung Humantech Paper Award (Feburary 2024)
- Encouragement Prize (Top 6%)
LG Display AI & Big Data Competition (December 2022)
- Excellence Prize
- Hosted by LG Display and Yonsei University
Yonsei Social Entrepreneurship Award (January 2020)
- Database-based LMS (learning management system) development
- Hosted by Yonsei University
Academic (Highest) Honors (Undergraduate, four times)
- 2016-1 (honors), 2019-1 (honors), 2019-2 (highest honors), 2021-1 (honors)
- GPA: 3.96/4.0 (4.17/4.3, Rank: 2/88)
Commendation from Jeju Provincial Police Agency (October 2017)
- Received while two years of compulsory military service
- Commendation from local police commissioner

Teaching Experience

Digital Systems Design and Lab. (Seoul National University)
- Teaching Assistant (Fall 2023)
Logic Circuit Design (CSI2111, Yonsei University)
- Teaching Assistant (Fall 2021)