News & Updates - System Software Lab, Korea University

Latest updates, paper acceptances, awards, and achievements from System Software Lab, Korea University.

Attended ACM APNet'23 (Sep. 2023)

Prof. Yang, Yeonho, Wonmi, and Prof. Yoo attended APNet23. We met our friends and research colleagues from Micrsoft Research Asia!

Attended ACM APNet'23 Attended ACM APNet'23 Full Image

SSLab launched! (Sep. 2023)

SSLab launched in September 2023. We highly seek enthusiastic and creative graduate students! 😃

SSLab launched! SSLab launched! Full Image

Eight papers accepted at KSC 2024! 🎉 (Nov. 2024)

Eight of our papers have been accepted at the Korea Software Congress 2024 (KSC 2024). Congratulations to all the authors! 👍
- 강민철, 고영훈, 유혁, 양경식, eBPF 기반 실시간/경량 GPU 워크로드 추적 기법, 2024 한국소프트웨어종합학술대회 (KSC2024) 🏆 Best paper award!
- 조휘주, 최원미, 최종인, 유혁, 양경식, 6G 네트워크 시스템 구축을 위한 O-RAN 및 NVIDIA AI Aerial 플랫폼 비교분석, 2024 한국소프트웨어종합학술대회 (KSC2024)
- 권도혁, 조휘주, 유연호, 유혁, 양경식, 네트워크 트래픽 사이즈 예측 모델의 경량화를 위한 양자화 기법 분석 및 비교, 2024 한국소프트웨어종합학술대회 (KSC2024)
- 박재림, 신창용, 양경식, 유혁, 양자화 기법의 거대 언어 모델 간 메모리 및 추론 속도 비교 분석, 2024 한국소프트웨어종합학술대회 (KSC2024)
- 안주영, 박선재, 최원미, 유혁, 양경식, 대규모 언어 모델 대상 분할-연합학습에서의 레이어별 통신 비용 분석, 2024 한국소프트웨어종합학술대회 (KSC2024) 🏆 Best paper presentation award!
- 최현목, 이현호, 유연호, 양경식, 유혁, 분산 딥러닝 학습에서 네트워크 대역폭과 모델 특성에 따른 학습 시간 변화 분석, 2024 한국소프트웨어종합학술대회 (KSC2024) 🏆 Best undergraduate paper award!
- 강준혁, 강민철, 고영훈, 유혁, 양경식, GPU 구성 요소의 성능 및 전력 소비에 대한 민감도 분석, 2024 한국소프트웨어종합학술대회 (KSC2024)
- 노태윤, 정승우, 유연호, 양경식, 유혁, 분할-연합학습의 데이터, 자원 이질성 극복을 위한 연구 동향 분석, 2024 한국소프트웨어종합학술대회 (KSC2024) 🏆 Best undergraduate paper award!

Eight papers accepted at KSC 2024! 🎉 Eight papers accepted at KSC 2024! 🎉 Full Image

Our lab won best poster award at AI Tech Day 2024! (Oct. 2024)

Our lab won the Best Poster Award at AI Tech Day 2024! We presented predictions for AI infrastructure related to model training, our federated learning technique, and predictive node placements on clouds. Special thanks to Jinwoo, Junseok, and Juyoung for their efforts!

Our lab won best poster award at AI Tech Day 2024! Our lab won best poster award at AI Tech Day 2024! Full Image

Seungwoo & Yeonho's paper accepted (Sep. 2024)

Our work on private blockchain configuration through prediction models has been accepted for publication in ICT Express. Great work!

Seungwoo & Yeonho's paper accepted Seungwoo & Yeonho's paper accepted Full Image

Wonmi & Yeonho's SCON accepted (Sep. 2024)

Our work on accelerating IoT container networking has been accepted for publication in the IEEE Internet of Things Journal. Great work!

Wonmi & Yeonho's SCON accepted Wonmi & Yeonho's SCON accepted Full Image

Invited talk on edge AI systems (Aug. 2024)

Prof. Blesson Varghese from the University of St Andrews gave an invited talk on edge AI systems. 🤖
- Title: Opportunities in Machine Learning Systems for the Edge
- Abstract: The premise of edge computing is to offer services on devices closer to the user than from the cloud. In this talk, I will discuss our experience of making machine learning (ML) work for the edge. I will discuss a range of techniques developed in my lab with my students that address computation and communication bottlenecks to make ML systems work on edge devices.

Invited talk on edge AI systems Invited talk on edge AI systems Full Image

Two domestic journal papers accepted (Aug. 2024)

Two papers for The Journal of the Korean Institute of Communications and Information Sciences have been accepted.
- An Analysis on Inference Time, Accuracy, Communication, and GPU Memory Usage for Inference Batch of Large Language Models
- Analysis of Workload Performance and Resource Consumption of Lightweight Secure Virtualization Platforms in IoT/Edge Computing
Congratulations! 🎉

Two domestic journal papers accepted Two domestic journal papers accepted Full Image

Wonmi and Junseok presented their papers (Jul. 2024)

Wonmi and Junseok presented their research papers at IEEE CLOUD 2024 in Shenzhen, China. Well done! 👏 Our members Seungwoo, Juyoung, Mincheol, and Yunkyung also attended to experience cutting-edge research.
Junseok Lee, Yeonho Yoo, Chuck Yoo, Gyeongsik Yang, "Predictive Placement of Geo-distributed Blockchain Nodes for Performance Guarantee" 🌐
Wonmi Choi, Juyoung Ahn, Yeonho Yoo, Chuck Yoo, Gyeongsik Yang, "Harmonia: Accurate Federated Learning with All-Inclusive Dataset" 🤖
Congratulations to all the authors for their remarkable contributions! 🥳📚

Wonmi and Junseok presented their papers Wonmi and Junseok presented their papers Full Image

Three domestic papers accepted (Jul. 2024)

Three research papers on domestic conferences in Jeju Island during the summer of 2024 have been accepted. 🎉Congratulations to our undergraduate interns and MS students!
- Expert 수와 배치 크기에 따른 Mixture of Experts 학습 성능 분석, 2024 한국컴퓨터종합학술대회 (KCC 2024), 조영준, 고영훈, 신창용, 유연호, 유혁, 양경식
- 분산 딥러닝 학습 프로세스와 운영체제 커널 간 인터페이스 병목 분석, 2024 한국컴퓨터종합학술대회 (KCC 2024), 강민철, 강준혁, 신창용, 유연호, 유혁, 양경식
- 허가형 블록체인을 위한 보안 컨테이너 활용 방안 연구, 2024년도 한국통신학회 하계종합학술발표회, 강윤경, 노태윤, 이준석, 최원미, 유연호, 양경식, 유혁

Three domestic papers accepted Three domestic papers accepted Full Image Three domestic papers accepted Three domestic papers accepted Full Image Three domestic papers accepted Three domestic papers accepted Full Image Three domestic papers accepted Three domestic papers accepted Full Image Three domestic papers accepted Three domestic papers accepted Full Image Three domestic papers accepted Three domestic papers accepted Full Image

Prof. Yang and Dr. Yoo attended MobiSys 2024 (June. 2024)

Prof. Yang and Dr. Yoo attended ACM MobiSys 2024. 🏛️ Prof. Yang gave a talk at the Mobile AI Systems workshop on the topic "Analysis of Containerized Networking Systems in IoT Devices." 📡

Prof. Yang and Dr. Yoo attended MobiSys 2024 Prof. Yang and Dr. Yoo attended MobiSys 2024 Full Image

Two papers accepted at IEEE CLOUD 2024 🌐 (May. 2024)

We are excited to announce that two papers have been accepted at the 2024 IEEE CLOUD in Shenzhen, China. 🎉 Congratulations to Junseok, Yeonho, Juyoung, and Wonmi for their outstanding work! 🤖 [more]

Two papers accepted at IEEE CLOUD 2024 🌐 Two papers accepted at IEEE CLOUD 2024 🌐 Full Image

AI system-infra project launched! 🧐 (May. 2024)

🚀 Research project for "consortium learning" system, toward secure and high-performance AI training in private domains, has been accepted! 🎉 A heartfelt thanks to the team members for their hard work and dedication to the proposal! 🙌🏼

AI system-infra project launched! 🧐 AI system-infra project launched! 🧐 Full Image

6G network system project initiated! 🛜 (May. 2024)

Research project for "end-to-end performance assurance for 6G networks" has been accepted 📡.Our team will investigate the 6G cloud part ☁️. This project will be carried out in collaboration with excellent research teams from SNU, KU, SKKU, UNIST, and Hanyang Univ 🎓.

6G network system project initiated! 🛜 6G network system project initiated! 🛜 Full Image

Dr. Yoo graduates! 🎓 (Mar. 2024)

🎉Congratulations to Dr. Yoo, who we worked with, on his PhD! He will be joining us as a postdoc to research interesting topics. We wish him the best of luck in his future endeavors!🍀

Dr. Yoo graduates! 🎓 Dr. Yoo graduates! 🎓 Full Image

Attended CSC 2024 (Jan. 2024)

Our team attended CSC 2024 (컴퓨터시스템 동계학술대회 2024)! Prof. Yang had a talk at the newcomer session. It was a great research experience and a time of fellowship ⛷️.

Attended CSC 2024 Attended CSC 2024 Full Image

Attended HSN 2024! (Jan. 2024)

Prof. Yang attended the 34th HSN 2024 and had a talk at the next-generation datacenter session!

Attended HSN 2024! Attended HSN 2024! Full Image

New M.S. student: Minuk 🎓 (Dec. 2025)

After completing his undergraduate internship, Minuk has officially started his M.S. program. Welcome onboard! 🙌⛵

New M.S. student: Minuk 🎓 New M.S. student: Minuk 🎓 Full Image

Four papers accepted at domestic conference! 🎁 (Dec. 2025)

Four of our papers have been accepted!
Congratulations to all! 🎊
- 남윤수, 조휘주, 최종인, 유혁, 양경식, 서버리스 워크로드의 QoS 보장을 위한 저지연 컨테이너 기법 비교분석, 2025 한국소프트웨어종합학술대회 (KSC 2025)
- 최종인, 조휘주, 남윤수, 정진우, 유혁, 양경식, NVIDIA DALI 실행 모드에 따른 JPEG 이미지 전처리 성능 비교 분석, 2025 한국소프트웨어종합학술대회 (KSC 2025)
- 이민하, 강민철, 유혁, 양경식, 서명 기반 비인가 BPF 프로그램 로드 차단 기법, 2025 한국소프트웨어종합학술대회 (KSC 2025)
- 김건민, 박선재, 최원미, 안주영, 유혁, 양경식, 엣지 장치에서의 LLM 학습 시 에너지 효율성 최적화 전략 비교분석, 2025 한국소프트웨어종합학술대회 (KSC 2025)

Four papers accepted at domestic conference! 🎁 Four papers accepted at domestic conference! 🎁 Full Image

Minuk selected as a next-generation engineer by IPESK (Sep. 2025)

Congratulations to Minuk on being named a Next-Generation Engineer by the Korea Institute for the Promotion of Engineering and Science Korea, IPESK. We look forward to seeing your continued research and future contributions to the engineering field.

Minuk selected as a next-generation engineer by IPESK Minuk selected as a next-generation engineer by IPESK Full Image

Sunjae received best presentation award! 🏆 (Aug. 2025)

Sunjae received the best presentation award for his paper, “Edge LLM Training Performance via Pipeline Parallelism: Understanding Communication and Micro-Batch Operations.” Big congratulations! 🎉

Sunjae received best presentation award! 🏆 Sunjae received best presentation award! 🏆 Full Image

Yeonho appointed as an assistant professor 👨‍🏫 (Aug. 2025)

We're thrilled to share that Dr. Yoo, our research professor, has been appointed as Assistant Professor at Dongguk University. Thank you for your dedication and hard work—wishing you continued success in this next chapter! 🤞

Yeonho appointed as an assistant professor 👨‍🏫 Yeonho appointed as an assistant professor 👨‍🏫 Full Image

Four papers accepted at domestic conferences! 🎉 (May. 2025)

Four of our papers have been accepted!
Congratulations to all the authors! 👍
- 신정윤, 신창용, 유혁, 양경식, 지리 분산 학습 환경에서 GPU 성능을 고려한 배치 크기 및 로컬 스텝 수 조정 기법, 2025 한국통신학회 하계종합학술발표회
- 박선재, 최원미, 안주영, 유혁, 양경식, 파이프라인 병렬화를 통한 엣지 LLM 학습 성능: 통신과 마이크로 배치 연산의 이해, 2025 한국컴퓨터종합학술대회 (KCC2025)
- 강준혁, 권도혁, 고영훈, 유혁, 양경식, 경량화된 머신러닝 모델의 추론 일관성 분석: Explainable AI 기반 정량 비교, 2025 한국컴퓨터종합학술대회 (KCC2025)
- 정진우, 한재훈, 강민철, 유혁, 양경식, 이기종 GPU 환경에서 Unified Virtual Memory 기반 LLM 주요 커널 실행 성능 분석, 2025 한국컴퓨터종합학술대회 (KCC2025)

Four papers accepted at domestic conferences! 🎉 Four papers accepted at domestic conferences! 🎉 Full Image

Two domestic journal papers accepted (May. 2025)

Two papers for The Journal of the Korean Institute of Communications and Information Sciences have been accepted.
- Wonmi Choi, Juyoung Ahn, Zhixiong Niu, Gyeongsik Yang, Chuck Yoo, Unveiling Kubernetes CNI: An In-Depth Analysis of Networking Performance and Resource Efficiency
- YunKyung Kang, Wonmi Choi, Gyeongsik Yang, Chuck Yoo, Performance and Resource Consumption Analysis of Hyperledger Fabric Based on Secure Containers
Congratulations! 🎉

Two domestic journal papers accepted Two domestic journal papers accepted Full Image

Yeonho's VALO accepted at Sigmetrics 2025 (Mar. 2025)

Yeonho’s paper on traffic splitting in datacenters, VALO, has been accepted at ACM Sigmetrics 2025! 📄✨VALO solves a traditional but critical inaccuracy problem of traffic splitting by its novel modeling approach. 🚀 Congratulations 🥳 to Yeonho on this impressive achievement!

Yeonho's VALO accepted at Sigmetrics 2025 Yeonho's VALO accepted at Sigmetrics 2025 Full Image

New graduate students: Jaerim and Sunjae 🎓 (Jan. 2025)

Welcome Jaerim Park (MS course) and Sunjae Park (PhD-MS integrated course) to our team!
Looking forward to exciting achievements together. Let’s make it a fantastic journey! 🚀✨ Welcome onboard! 🙌🎉

New graduate students: Jaerim and Sunjae 🎓 New graduate students: Jaerim and Sunjae 🎓 Full Image

Four papers accepted at EMNLP 2026! 🎊 (Aug. 2026)

Four of our papers have been accepted at the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026) to be held in Budapest, Hungary! 🚀

Congratulations to all the authors! 👏
- Seungwoo Jung, Dohyeok Kwon, Seungmin Cha, Junseok Lee, Yeonho Yoo, Chuck Yoo, Gyeongsik Yang, "Residual Sparsification via Output Importance for Compressing Mixture-of-Experts LLMs" (Main Track)
- Younghun Go, Jaehoon Han, Changyong Shin, Chuck Yoo, Gyeongsik Yang, "Enabling KV Caching of Shared Prefix for Diffusion Language Models" (Main Track)
- Minchul Kang, Changyong Shin, Younghun Go, Hyunho Lee, Jinwoo Jeong, Chuck Yoo, Gyeongsik Yang, "Parallelism Strategy Chaining for Fast Training Convergence" (Main Track)
- Changyong Shin, Jaerim Park, Minchul Kang, Younghun Go, Zhixiong Niu, Yongqiang Xiong, Gyeongsik Yang, Chuck Yoo, "GeoMesh: Workload-Balanced and Sign-Compressed Geo-Distributed LLM Training" (Findings)

Four papers accepted at EMNLP 2026! 🎊 Four papers accepted at EMNLP 2026! 🎊 Full Image

Three papers accepted at MASCOTS 2026! 🎊 (Aug. 2026)

Three of our papers have been accepted for the 34th IEEE International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS 2026, Acceptance rate ~23%) to be held in Genova, Italy! 🚀

Congratulations to all the authors! 👏
- Wonmi Choi, Minuk Park, Zhixiong Niu, Yongqiang Xiong, Chuck Yoo, Gyeongsik Yang, "Not All AI Agents Are Equal: Characterizing Resource and Performance Dynamics"
- Wonmi Choi, Sunjae Park, Dohyeok Kwon, Zhixiong Niu, Yeonho Yoo, Chuck Yoo, Gyeongsik Yang, "Xronos: Heterogeneity-Aware Tensor Parallelism for Collaborative LLM Fine-Tuning on Edge CPUs"
- Yeonho Yoo, Hyunho Lee, Hyunmok Choi, Chuck Yoo, Gyeongsik Yang, "Accurate Simulation of Distributed Training Jobs with Network Contention Modeling"

Three papers accepted at MASCOTS 2026! 🎊 Three papers accepted at MASCOTS 2026! 🎊 Full Image

Younghun's Lazer nominated for Best Paper Candidate at DAC 2026! 🏆 (Jul. 2026)

Our paper, "Making Sense of Job Preemption for Distributed Deep Learning Acceleration", has been nominated as a Best Paper Candidate at the 63rd ACM/IEEE Design Automation Conference (DAC 2026)!

Out of 2,443 paper submissions, only 13 papers (0.5%) were selected as Best Paper Candidates. 👏

Congratulations to Younghun and all the co-authors! 🚀

Younghun's Lazer nominated for Best Paper Candidate at DAC 2026! 🏆 Younghun's Lazer nominated for Best Paper Candidate at DAC 2026! 🏆 Full Image

Yeonho and Junseok's Xalute accepted at Scientific Reports! 🎊 (Jul. 2026)

Our paper, "Unified Digital Biomarker Platform for Wearable Devices Assuring Integrity, Heterogeneity, and Scalability", has been accepted for publication in Scientific Reports. Congratulations to Yeonho, Junseok, Seungwoo, and all the authors! 🚀

Yeonho and Junseok's Xalute accepted at Scientific Reports! 🎊 Yeonho and Junseok's Xalute accepted at Scientific Reports! 🎊 Full Image

Best paper award at KCC 2026! 🏆 (Jun. 2026)

Jongin received the Best Paper Award (최우수논문상) at the 2026 Korea Computer Congress (KCC 2026) for the paper titled "6G 오프로드 응용을 위한 추론 예측 기법 분석".

Congratulations to all the authors! 🎉

Best paper award at KCC 2026! 🏆 Best paper award at KCC 2026! 🏆 Full Image

Four papers accepted at domestic conference! 🎁 (May. 2026)

Four of our papers have been accepted!
Congratulations to all the authors! 🚀
- 박주영, 이준석, 정승우, 유혁, 양경식, 인종에 따른 심전도 파운데이션 모델 성능 분석, 2026 한국컴퓨터종합학술대회 (KCC 2026)
- 최종인, 조휘주, 남윤수, 유혁, 양경식, 6G 오프로드 응용을 위한 추론 예측 기법 분석, 2026 한국컴퓨터종합학술대회 (KCC 2026)
- 박민욱, 최원미, 유혁, 양경식, 에이전트 시스템에서의 LLM 추론 속도에 따른 CPU 바운드 도구 실행 특성 분석, 2026 한국컴퓨터종합학술대회 (KCC 2026)
- 권도혁, 최원미, 유혁, 양경식, 파이프라인 병렬화 기반 엣지 LLM 미세조정에서 GPU 거버너의 영향 분석, 2026 한국컴퓨터종합학술대회 (KCC 2026)

Four papers accepted at domestic conference! 🎁 Four papers accepted at domestic conference! 🎁 Full Image

Younghun's Lazer accepted at DAC 2026! 🎊 (Feb. 2026)

Our paper "Making Sense of Job Preemption for Distributed Deep Learning Acceleration" has been accepted for publication in the DAC 2026. Congratulations to Younghun, Changyong, Minchul and all the authors! 🚀

Younghun's Lazer accepted at DAC 2026! 🎊 Younghun's Lazer accepted at DAC 2026! 🎊 Full Image

Excellence award at KSC 2025 paper competition! 🏆 (Jan. 2026)

Geonmin, Sunjae, Wonmi, and Juyoung received the Excellence Award in the Undergraduate Paper Competition at the 2025 Korea Software Congress (KSC 2025) for the paper titled "Comparative Analysis of Energy Efficiency Optimization Strategies for LLM Training on Edge Devices". Congratulations to all the authors! 🎉

Excellence award at KSC 2025 paper competition! 🏆 Excellence award at KSC 2025 paper competition! 🏆 Full Image

Changyong's TensorShare accepted at FGCS! 🎊 (Feb. 2026)

Our paper titled "Prediction-based GPU Sharing for Distributed Training" has been accepted for publication in Future Generation Computer Systems (FGCS). Congratulations to Changyong and all the authors! 🚀

Changyong's TensorShare accepted at FGCS! 🎊 Changyong's TensorShare accepted at FGCS! 🎊 Full Image