Yanpeng Yu

Yanpeng Yu

PhD student @ Yale University

Department of Computer Science

Yale University

Biography

Download my resumé.

I received my Ph.D. in Computer Science from Yale University, where I was advised by Professor Anurag Khandelwal and Professor Lin Zhong. My research focuses on designing efficient cache-coherence and synchronization protocols for large-scale shared-memory systems.

During my Ph.D., I spent two summers as a research intern with the Architecture Research Group at NVIDIA, where I worked on systems and architectures for Large Language Models (LLMs).

My work was named an IEEE Micro Top Pick in Computer Architecture (2025) and received a Distinguished Artifact Award at ISCA 2025. Techniques from my research have been adopted in NVIDIA’s Vera Rubin platform.

Before Yale, I completed my undergraduate studies at Peking University.

Interests
  • Memory Disaggregation
  • Cache Coherence Protocols
  • Synchronization Protocols
  • Architecture/System for ML
Education
  • Ph.D. in Computer Science, 2021-Present

    Yale University

  • M.Sc. in Computer Science, 2021-2024

    Yale University

  • B.Sc. in Computer Science and Technology, 2017-2021

    Peking University

Recent Publications

(2026). Efficient and Scalable Synchronization via Generalized Cache Coherence. OSDI'26.

(2025). Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens. In Submission.

PDF

(2025). CORD: Low-Latency, Bandwidth-Efficient and Scalable Release Consistency via Directory Ordering. ISCA'25 Distinguished Artifact Award. Selected for IEEE Micro’s Top Picks in Computer Architecture in 2025.

PDF Code

(2023). An Efficient Data Structure for Dynamic Graph on GPUs. TKDE'23.

PDF

(2021). MIND: In-Network Memory Management for Disaggregated Data Centers. SOSP'21.

PDF Code

Contact