Jaehyeong Park

I am a PhD candidate in Electrical Engineering in the INA Lab at KAIST, advised by Prof. Dongsu Han. I received my M.S. degree in Electrical Engineering from KAIST, where I was also advised by Prof. Dongsu Han. Also, I have received BS degree in Electrical Engineering at KAIST.

My research focuses on tensor program optimization for accelerating deep learning models. Currently, I am developing a multi-level superoptimizer for tensor programs, aiming to optimize tensor operations to better align with GPU characteristics and enhance computation speed. Previously, I worked on various topics, including the automatic generation of audio descriptions for movies and resource optimization in SLO-oriented microservices.

Publications

ASPLOS

Trinity: Three-Dimensional Tensor Program Optimization via Tile-level Equality Saturation

Jaehyeong Park, Youngchan Kim, Haechan An, Gieun Jeong, Jeehoon Kang, and Dongsu Han

ACM International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2026
(To appear)

WACV

NarrAD: Automatic Generation of Audio Descriptions for Movies with Rich Narrative Context

Jaehyeong Park, Juncheol Ye, Seungkook Lee, Hyun W. Ka, and Dongsu Han

In Proceedings of the IEEE/CVF Winter Conference, WACV 2025
(Oral, Top 8.3% of submissions)

SIGCOMM

TopFull: An Adaptive Top-Down Overload Control for SLO-Oriented Microservices

Jinwoo Park, Jaehyeong Park, Youngmok Jung, Hwijoon Lim, Hyunho Yeo, and Dongsu Han

ACM Special Interest Group on Data Communication, SIGCOMM 2024

Abs HTML

Microservice has become a de facto standard for building large-scale cloud applications. Overload control is essential in preventing microservice failures and maintaining system performance under overloads. Although several approaches have been proposed, they are limited to mitigating overload at individual microservices lacking assessments over the interdependent microservices and APIs. This paper presents TopFull, a holistic overload control framework for microservices that leverage global coordination to maximize the throughput that meet service level objectives (i.e., goodput). TopFull (a) dynamically cluster APIs according to the dependency with overloaded microservices, (b) choose APIs to load-control among those that are in contending relationships, and (c) take actions from RL agents which adaptively adjust the admitted rates of the APIs to maximize the goodput. Our experiments on various open-source benchmarks demonstrate that TopFull significantly increases the goodput under overload scenarios, outperforming DAGOR by 1.82x and Breakwater by 2.26x. Furthermore, Kubernetes autoscaler with TopFull serves up to 3.91x more requests under traffic surge and tolerates traffic spikes with up to 57% fewer resources than the Kubernetes autoscaler standalone.