Nanqinqin Li

Nanqinqin Li

Soccer fan & Gamer, and Systems Researcher on the side

Bio

My name is Nanqinqin Li (Chinese: 李南沁沁). I am a Research Scientist at ByteDance, working on the next-generation memory architectures for AI and agentic workloads. I got my Ph.D. in CS from Princeton University, advised by Prof. Michael Freedman on systems research in Princeton SNS Group. I also worked closely with Prof. Amit Levy, Prof. Wyatt Lloyd, and Prof. Asaf Cidon(Columbia).

I received my M.S. degree from the University of Chicago advised by Prof. Haryadi Gunawi on combating the notorious tail latency problem of solid-state drives. I got my B.Eng degree from Beijing University of Posts and Telecommunications.

Research interests

I am broadly interested in building fast, robust systems with emerging memory and storage technologies (e.g., CXL). My current project focuses on reducing CPU idle time when CXL introduces higher memory access latency.

Publications

Peer-reviewed

  • Harvesting Sub-Microsecond CXL Memory Stalls with LiteSwitch
    Nanqinqin Li, Yuhong Zhong, Asaf Cidon, and Michael J. Freedman.
    The 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2026).
    [bib]

  • Speculative Recovery: Cheap, Highly Available Fault Tolerance with Disaggregated Storage
    Nanqinqin Li, Anja Kalaba, Michael J. Freedman, Wyatt Lloyd, and Amit Levy.
    The 2022 USENIX Annual Technical Conference (ATC 2022).
    [paper][talk][slides][code][bib]

  • [Best Paper Award] Fantastic SSD Internals and How to Learn and Use Them
    Nanqinqin Li, Mingzhe Hao, Huaicheng Li, Xing Lin, Tim Emami, and Haryadi S. Gunawi.
    The 15th ACM International Systems and Storage Conference (SYSTOR 2022).
    [paper][talk][code][bib]

  • LinnOS: Predictability on Unpredictable Flash Storage with a Light Neural Network
    Mingzhe Hao, Levent Toksoz, Nanqinqin Li, Edward Edberg, Henry Hoffmann, and Haryadi S. Gunawi.
    The 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2020).
    [paper][talk (Mingzhe Hao)][slides][code][bib]

arXiv

  • STORREP: Storage Research Experiment Patterns on Chameleon Cloud and Trovi
    Ray AO Sinurat, Yuyang Huang, Nanqinqin Li, Mark Powers, Michael Sherman, Kate Keahey, and Haryadi S. Gunawi.
    [paper][bib]

  • How Low Can You Go? Practical cold-start performance limits in FaaS
    Yue Tan, David Liu, Nanqinqin Li, and Amit Levy.
    [paper][code][bib]

Services

  • Artifact Evaluation PC - OSDI 2026

Teaching

  • Princeton University - COS 461, 418/518, 126, 217, 316