- About Us
- Welcome Message
- Vision
- Brief History
- Staff
- Contact Us
- Faculty / Research
- Faculty
- Research Labs
- Academics
- Curriculum
- Course Guide
- Graduate School
- Graduation Project
- Admissions
- Admissions
- Academic Affairs
- Academic Calendar
- Scholarships
- Graduation Requirements
- Student Clubs
- Academic Information
- Facility Reservations
- Community / News
- NEWS
- Notices
- Event
- Job Openings
- Forms
- KU Q&A
Community / News
NEWS
Korea University Ph.D. Student Soonjae Hwang and Professor Jeongseob Ahn`s Research Team Achieves Acceptance of a Paper at OSDI 2026, the Premier Conference in Computer Systems
관리자 2026.06.16 Views 25
▲ Prof. Jeongseob Ahn of Korea University (left) and Soonjae Hwang, Ph.D. student at Korea University (right)
A paper by Ph.D. student Soonjae Hwang and Professor Jeongseob Ahn of the Computer Systems Laboratory at Korea University has been accepted to OSDI (USENIX Symposium on Operating Systems Design and Implementation) 2026, one of the most prestigious international conferences in the field of computer systems.
This marks the first time that a paper from Korea University has been accepted to OSDI, demonstrating the university's world-class research competitiveness in operating systems and computer systems.
The research team developed a novel scheduling technique that reduces resource waste when serving large language models (LLMs) across multiple GPUs. Their approach achieved higher throughput than the current industry-standard tensor parallelism while using only conventional GPU servers, without requiring expensive high-speed interconnects such as NVLink.
This achievement is significant because it opens the door to lowering the operational costs of LLM services while improving resource efficiency.
-Paper: https://www.usenix.org/conference/osdi26/presentation/hwang (to be publicly available on July 15)
This marks the first time that a paper from Korea University has been accepted to OSDI, demonstrating the university's world-class research competitiveness in operating systems and computer systems.
The research team developed a novel scheduling technique that reduces resource waste when serving large language models (LLMs) across multiple GPUs. Their approach achieved higher throughput than the current industry-standard tensor parallelism while using only conventional GPU servers, without requiring expensive high-speed interconnects such as NVLink.
This achievement is significant because it opens the door to lowering the operational costs of LLM services while improving resource efficiency.
-Paper: https://www.usenix.org/conference/osdi26/presentation/hwang (to be publicly available on July 15)
File
Prev
2026.06.15
Next
No next post.
Engineering Building #412, College of Engineering, 145 Anam-ro, Seongbuk-gu, Seoul, 02841, KoreaTEL : 02-3290-3210, 3211, 3213, 3214, 4592FAX : 02-921-0544
Copyrights ⓒ 2021 KOREA University. All Rights Reserved.
- Family Site