About

I am a Ph.D. student at UC Berkeley, advised by Prof. Ion Stoica. I am interested in building practical, flexible, and high-performance software systems for emerging applications such as large language models.

woosuk.kwon[at]berkeley.edu

Interests
  • Machine Learning Systems
  • Cloud Computing
Education

  • Ph.D. in Computer Science

    UC Berkeley, 2021 - Present

  • B.S. in Computer Science and Mathematics

    Seoul National University, 2015 - 2021

Current Projects

vLLM
A high-throughput and memory-efficient inference and serving engine for LLMs.
GitHub Blog
vLLM
SkyPilot
A framework for easily and cost effectively running machine learning workloads on any cloud.
GitHub Blog
SkyPilot

Publications

Efficient Memory Management for Large Language Model Serving with PagedAttention
Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph E. Gonzalez, Hao Zhang, Ion Stoica
SOSP 2023
PDF Code
SkyPilot: An Intercloud Broker for Sky Computing
Zongheng Yang, Zhanghao Wu, Michael Luo, Wei-Lin Chiang, Romil Bhardwaj, Woosuk Kwon, Siyuan Zhuang, Frank Sifei Luan, Gautam Mittal, Scott Shenker, Ion Stoica
NSDI 2023
PDF Code
A Fast Post-Training Pruning Framework for Transformers
Woosuk Kwon, Sehoon Kim, Michael W. Mahoney, Joseph Hassoun, Kurt Keutzer, Amir Gholami
NeurIPS 2022
PDF Code
Learned Token Pruning for Transformers
Sehoon Kim, Sheng Shen, David Thorsley, Amir Gholami, Woosuk Kwon, Joseph Hassoun, Kurt Keutzer
KDD 2022
PDF Code
Nimble: Lightweight and Parallel GPU Task Scheduling for Deep Learning
Woosuk Kwon, Gyeong-In Yu, Eunji Jeong, Byung-Gon Chun
NeurIPS 2020 (Spotlight)
PDF Code
Graphene: Strong yet Lightweight Row Hammer Protection
Yeonhong Park, Woosuk Kwon, Eojin Lee, Tae Jun Ham, Jung Ho Ahn, Jae W. Lee
MICRO 2020 (IEEE Micro Top Picks Honorable Mention)
PDF