Gan Fang 方淦

About Me

I am a Ph.D. candidate in Computer Science at Purdue University, advised by Dr. Changhee Jung. My research focuses on computer architecture, with an emphasis on designing high-performance and energy-efficient CPUs and GPUs, Large Language Model (LLM) acceleration, compiler-architecture co-design, and cache hierarchies.

Recognitions

• 2026-2027 Bilsland Dissertation Fellowship
• ISCA 2025 Best Paper Honorable Mention

Publications

• EMSOFT 2026
  Achieving Speculative Intermittent Computation on Commodity Energy Harvesting Hardware

  Yida Zhang, Shao-Yu Huang, Byounguk Min, Andrew Inho Park, Gan Fang, Jongouk Choi and Changhee Jung
• IEEE Computer
  Speculation under Intermittence: A Guide for Computer Architects

  Gan Fang and Changhee Jung
• HPCA 2026
  Intermittence-Aware Cache Compression

  Gan Fang, Jianping Zeng, Yuchen Zhou, and Changhee Jung
• ISCA 2025
  Rethinking Prefetching for Intermittent Computing

  Gan Fang, Jianping Zeng, Aditya Gupta, and Changhee Jung

  Best Paper Honorable Mention
• HPCA 2025
  Rethinking Dead Block Prediction for Intermittent Computing

  Gan Fang and Changhee Jung
• ICCAD 2024
  Hybrid Power Failure Recovery for Intermittent Computing

  Gan Fang, Jongouk Choi, and Changhee Jung

Experience

• May – Aug 2026
  Computer Architecture Research Intern Futurewei Technologies
• May – Aug 2025
  SoC Modeling Intern Samsung Research America

Meow

Dax is a three-year-old British Shorthair who joined our family in the fall of 2023. He's also my dedicated late-night companion, faithfully helping me debug ideas and polish papers along the way.