Byungsoo Oh

byungsoo_pic.jpg

CS PhD Student @ Cornell University

Hi! I am a first-year PhD student in Computer Science at Cornell University. My research interests are in improving machine learning systems and cloud computing in terms of performance, resource-efficiency, and usability.

Previously, I worked at Samsung Research for 4.5 years, where I conducted research on accelerating systems for ML training. I also developed and operated an ML cloud system built on a large-scale GPU infrastructure. Prior to that, I earned my M.S. in Computer Science at KAIST. I received my B.S. in Computer Science & Engineering from Sogang University with Summa Cum Laude.

News

Aug 21, 2024 Started my PhD studies in Cornell CS.
Jun 07, 2024 🛫 Awarded student grant to attend USENIX ATC 2024 in Santa Clara!
May 01, 2024 🎉 Our paper has been accepted to USENIX ATC 2024!
Dec 19, 2022 🎉 Our paper has been accepted to VLDB 2023!

Publications

  1. USENIX ATC
    Metis: Fast Automatic Distributed Training on Heterogeneous GPUs
    Taegeon Um*,  Byungsoo Oh* , Minyoung Kang*, Woo-Yeon Lee, Goeun Kim, Dongseob Kim, Youngtaek Kim, Mohd Muzzammil, and Myeongjae Jeon (* = Equal Contribution)
    In USENIX Annual Technical Conference (USENIX ATC) , 2024
    HTML
  2. VLDB
    FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline
    Taegeon Um,  Byungsoo Oh , Byeongchan Seo, Minhyeok Kweun, Goeun Kim, and Woo-Yeon Lee
    In International Conference on Very Large Data Bases (VLDB) , 2023
    HTML Code
  3. IPDPS
    PokéMem: Taming Wild Memory Consumers in Apache Spark
    Minhyeok Kweun, Goeun Kim,  Byungsoo Oh , Seongho Jung, Taegeon Um, and Woo-Yeon Lee
    In IEEE International Parallel and Distributed Processing Symposium (IPDPS) , 2022
    HTML
  4. IJCNN
    DAPAS: Denoising Autoencoder to Prevent Adversarial attack in Semantic Segmentation
    Seungju Cho, Tae Joon Jun,  Byungsoo Oh , and Daeyoung Kim
    In International Joint Conference on Neural Networks (IJCNN) , 2020
    HTML
  5. Middleware
    Serverless-Enabled Permissioned Blockchain for Elastic Transaction Processing
    Byungsoo Oh , and Daeyoung Kim
    In ACM/IFIP International Middleware Conference (Middleware), Demos and Posters , 2019
    HTML Poster
  6. SMC
    Enhancing Trust of Supply Chain Using Blockchain Platform with Robust Data Model and Verification Mechanism
    Byungsoo Oh , Tae Joon Jun, Wondeuk Yoon, Yunho Lee, Sangtae Kim, and Daeyoung Kim
    In IEEE International Conference on Systems, Man and Cybernetics (SMC) , 2019
    HTML Slides