闪英迪｜清华大学｜学者主页

期刊论文

2026

Surviving Partial Rank Failures in Wide Expert-Parallel MoE Inference

Xun Sun, Shao-Yuan Chen, Ping-Chuan Ma, Yueqiang Chen, Zi-Wei Yuan, Zhanhao Cao, Han Han, Shangming Cai, Teng Ma, Xuchun Shang, Xin-Peng Zhao, Ke Yang, J. Wei, Lian-Zhi Lin, Yujia Liu, Feng Ren, Haoran Hu, Cheng Wan, Yingdi Shan, Yong-Wei Wu, Ming-Xing Zhang

arXiv.org

DOI

2026

SplineFS: Adaptive Learned Directory Indexing for Memory-Constrained Filesystems (ACM ATC '26 Artifact)

Jianqiang Zeng, Senfeng Sun, Chengying Huan, Peng Wang, Ke Zhou, Yingwei Luo, Xiaolin Wang, Zhenlin Wang, Diyu Zhou, Yingdi Shan, Jie Zhang

Zenodo (CERN European Organization for Nuclear Research)

DOI

2025

Scalio: Scaling up DPU-based JBOF Key-value Store with NVMe-oF Target Offload

Xun Sun, Ming-Xing Zhang, Yingdi Shan, Kang Chen, Jinlei Jiang, Yong-Wei Wu

USENIX Symposium on Operating Systems Design and Implementation

2025

DSA-2LM: A CPU-Free Tiered Memory Architecture with Intel DSA

Ruilin Liu, Teng Ma, Ming-Xing Zhang, Jia-Liang Huang, Yingdi Shan, Zheng Liu, Lingfeng Xiang, Zhen Lin, Hui Lu, Jia Rao, Kang Chen, Yong-Wei Wu

USENIX Annual Technical Conference

2025

Seer: Online Context Learning for Fast Synchronous LLM Reinforcement Learning

Ruoyu Qin, Wei-Ran He, Weixiao Huang, Yang-Kun Zhang, Yi-Kai Zhao, Bo Pang, Xinran Xu, Yingdi Shan, Yong-Wei Wu, Ming-Xing Zhang

arXiv.org

DOI

2025

OOCC: One-Round Optimistic Concurrency Control for Read-Only Disaggregated Transactions

Hao Wu, Ming-Xing Zhang, Kang Chen, Xia Liao, Yingdi Shan, Yong-Wei Wu

IEEE International Conference on Data Engineering

DOI

2025

Scaling Asynchronous Graph Query Processing via Partitioned Stateful Traversal Machines

Shao-Yuan Chen, Hongtao Chen, Shaonan Ma, Yajie Qin, Zheng Wang, Weiyu Xie, Ming-Xing Zhang, Kang Chen, Xia Liao, Yingdi Shan, Jinlei Jiang, Yong-Wei Wu

IEEE International Conference on Data Engineering

DOI

2025

TrEnv-X: Transparently Share Serverless Execution Environments Across Different Functions and Nodes

Jia-Liang Huang, Teng Ma, Zheng Liu, Si-Xing Lin, Kang Chen, Jinlei Jiang, Xia Liao, Yingdi Shan, Yong-Wei Wu, Ning Zhang, Mengting Lu, Tao Ma, Hai-Feng Gong, Mingxing Zhang

ACM Transactions on Computer Systems

DOI

2025

Accelerating Stream Processing Engines via Hardware Offloading

Zhengyan Guo, Ming-Xing Zhang, Yingdi Shan, Kang Chen, Jinlei Jiang, Yong-Wei Wu

Proc. ACM Manag. Data

DOI

2025

Efficient Graph-Based Approximate Nearest Neighbor Search Achieving: Low Latency Without Throughput Loss

Jingjia Luo, Ming-Xing Zhang, Kang Chen, Xia Liao, Yingdi Shan, Jinlei Jiang, Yong-Wei Wu

arXiv.org

DOI

2024

Efficient Heterogeneous Large Language Model Decoding with Model-Attention Disaggregation

Shao-Yuan Chen, Wencong Xiao, Yu-Tong Lin, Ming-Xing Zhang, Yingdi Shan, Jinlei Jiang, Kang Chen, Yong-Wei Wu

2024

TrEnv: Transparently Share Serverless Execution Environments Across Different Functions and Nodes

Jia-Liang Huang, Mingxing Zhang, Teng Ma, Zheng Liu, Si-Xing Lin, Kang Chen, Jinlei Jiang, Xia Liao, Yingdi Shan, Ning Zhang, Mengting Lu, Tao Ma, Hai-Feng Gong, Yong-Wei Wu

Symposium on Operating Systems Principles

DOI

2021

Geometric Partitioning: Explore the Boundary of Optimal Erasure Code Repair

Yingdi Shan, Kang Chen, Tuoyu Gong, Li-Dong Zhou, Tai Zhou, Yong-Wei Wu

Symposium on Operating Systems Principles

DOI

预印本

2026

AFD-Ledger: Deployment Provisioning for Attention--FFN Disaggregation

Chengyu Qiu, Xiao Fu, Fengcun Li, Yulei Qian, Yuchen Xie, Xunliang Cai, Yingdi Shan, Yongwei Wu, Mingxing Zhang

arXiv (Cornell University)

DOI

闪英迪

精选作品

期刊论文

Surviving Partial Rank Failures in Wide Expert-Parallel MoE Inference

SplineFS: Adaptive Learned Directory Indexing for Memory-Constrained Filesystems (ACM ATC '26 Artifact)

Scalio: Scaling up DPU-based JBOF Key-value Store with NVMe-oF Target Offload

DSA-2LM: A CPU-Free Tiered Memory Architecture with Intel DSA

Seer: Online Context Learning for Fast Synchronous LLM Reinforcement Learning

OOCC: One-Round Optimistic Concurrency Control for Read-Only Disaggregated Transactions

Scaling Asynchronous Graph Query Processing via Partitioned Stateful Traversal Machines

TrEnv-X: Transparently Share Serverless Execution Environments Across Different Functions and Nodes

Accelerating Stream Processing Engines via Hardware Offloading

Efficient Graph-Based Approximate Nearest Neighbor Search Achieving: Low Latency Without Throughput Loss

Efficient Heterogeneous Large Language Model Decoding with Model-Attention Disaggregation

TrEnv: Transparently Share Serverless Execution Environments Across Different Functions and Nodes

Geometric Partitioning: Explore the Boundary of Optimal Erasure Code Repair

预印本

AFD-Ledger: Deployment Provisioning for Attention--FFN Disaggregation

研究方向