Shizhen Zhao | 赵世振

Shizhen Zhao (赵世振)

教授 · 上海交通大学

📧 shizhenzhao at sjtu.edu.cn

上海交大本科（信息工程 + 数学），普渡大学 ECE 博士（导师 Xiaojun Lin, IEEE Fellow）。

2015–2019 年任谷歌总部数据中心网络架构组软件工程师。

2019 年加入上海交大任教，现为教授、博士生导师。

入选教育部青年长江学者 | 上海市海外高层次人才计划

2023–2025 年任国家自然科学基金委信息中心流动项目主任

📢 招生信息

🔬 博士后：常年招聘，有产业项目支撑，待遇优厚，欢迎邮件联系

🎓 2028年入学博士生：欢迎申请（2027年名额已满）

📚 2027年入学硕士生：要求入围上海交通大学计算机学院/集成电路学院优秀大学生夏令营（优营）

感兴趣的同学请发送简历至 📧 shizhenzhao at sjtu.edu.cn

研究方向

研究围绕光电混合智算集群与AI for System Research两条主线展开：

光电混合智算集群：从光交换机到底层协议的全栈设计，涵盖快切换 OCS 与 Switch-NIC 协同架构、光电混合数据中心拓扑与路由、RDMA 可靠传输与集合通信优化、弹性训练与故障恢复等，构建面向万卡级 AI 集群的高性能网络基础设施。

AI for System Research：利用机器学习与可解释 AI 解决网络诊断与优化问题，包括系统知识库构建、系统瓶颈自定位与自优化、AI 智能运维（自动异常检测与根因定位），以及 AI Agent 从零构建完整可部署系统的工程能力探索。

相关成果发表在 SIGCOMM（上海交大首篇）、NSDI、EuroSys、TON 等国际顶级会议与期刊。

主持科研项目

国家自然科学基金联合重点项目 · "面向自智网络的自主路由架构与关键技术研究"

2026.01 – 2029.12 | 项目批准号：U25B2039 | 258万元

国家自然科学基金面上项目 · "面向AI分布式训练的光电混合数据中心网络"

2023.01 – 2026.12 | 项目批准号：62272292 | 55万元

国家自然科学基金青年项目 · "数据中心低频动态拓扑结构设计与优化"

2020.01 – 2022.12 | 项目批准号：61902246 | 27万元

Selected Publications

Shengkai Lin, Kairui Zhou, Hongtao Zhang, Yibo Wu, Yi Pan, Yihan Yang, Qinwei Yang, Wei Zhang, Arvind Krishnamurthy, Shizhen Zhao, "SHIFT: Cross-NIC Fault Tolerance for RDMA Networks," in EuroSys, Rabat, Morocco, April, 2027. (Corresponding Author)
Ximeng Liu, Yingming Mao, Zhuoran Liu, Jinhui Ren, Shuyao Qi, Xinchi Han, Siyuan Feng, Peirui Cao, Yatao Li, Shizhen Zhao, Xinbing Wang, "AICCL: Pushing the Quality-Scalability-Flexibility Frontier in Collective Communication via Learning," in NSDI, Providence, RI, USA, May, 2027. (Corresponding Author)
Ximeng Liu, Zhuoran Liu, Yingming Mao, Yatao Li, Shizhen Zhao, Xinbing Wang, "Geminet: Learning the Duality-based Topology-Agnostic Update Operator for Lightweight Traffic Engineering in Changing Topologies," in NSDI, Santa Clara, CA, USA, March, 2026. (Corresponding Author)
Weihao Jiang, Wenli Xiao, Yuqing Yang, Peirui Cao, Shizhen Zhao, "Orderlock: A New Type of Deadlock and its Implications on High Performance Network Protocol Design," in SIGCOMM, Coimbra, Portugal, September, 2025. (Corresponding Author)
Shengkai Lin, Shizhen Zhao, Peirui Cao, Xinchi Han, Quan Tian, Wenfeng Liu, Qi Wu, Donghai Han, Xinbing Wang, "ONCache: A Cache-Based Low-Overhead Container Overlay Network," in NSDI, Philadelphia, PA, USA, April, 2025. (Corresponding Author)
Xiwei Wu, Yueyang Feng, Tianyi Huang, Xiaoyang Lu, Shengkai Lin, Lihan Xie, Shizhen Zhao, Qinxiang Cao, "VEP: A Two-stage Verification Toolchain for Full eBPF Programmability," in NSDI, Philadelphia, PA, USA, April, 2025. (Co-Corresponding Author)
Ximeng Liu, Shizhen Zhao, Yong Cui, Xinbing Wang, "FIGRET: Fine-Grained Robustness-Enhanced Traffic Engineering," in SIGCOMM, Sydney, NSW, Australia, August, 2024. (Corresponding Author)
Cong Liang, Xiangli Song, Jing Cheng, Mowei Wang, Yashe Liu, Zhenhua Liu, Shizhen Zhao, Yong Cui, "NegotiaToR: Towards A Simple Yet Effective On-demand Reconfigurable Datacenter Network," in SIGCOMM, Sydney, NSW, Australia, August, 2024.
Shizhen Zhao, Qizhou Zhang, Peirui Cao, Xiao Zhang, Xinbing Wang, Chenghu Zhou, "Flattened Clos: Designing High-performance Deadlock-free Expander Data Center Networks Using Graph Contraction," in NSDI, Boston, MA, USA, April, 2023.
Shizhen Zhao, Peirui Cao, Xinbing Wang, "Understanding the Performance Guarantee of Physical Topology Design for Optical Circuit Switched Data Centers," in SIGMETRICS, Mumbai, India, June, 2022.
Shizhen Zhao, Rui Wang, Junlan Zhou, Joon Ong, Jeffery C. Mogul, Amin Vahdat, "Minimal Rewiring: Efficient Live Expansion for Clos Data Center Networks," in NSDI, Boston, MA, USA, February, 2019. (Technical Report)
Shizhen Zhao, Ximeng Liu, Tianyu Zhu, Xinbing Wang, "Pulse+: DetNet Routing Under Delay-diff Constraint," in IEEE/ACM Transactions on Networking, Vol. 33, No. 6, December, 2025. (Technical Report)
Wenli Xiao, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "Segment EDF: A Scheduling Policy with Tight Deterministic Latency under Multi-hop Networks," in IEEE/ACM Transactions on Networking, Vol. 33, No. 1, February, 2024. (Corresponding Author)
Peirui Cao, Shizhen Zhao, Dai Zhang, Zhuotao Liu, Mingwei Xu, Min Yee Teh, Yunzhuo Liu, Xinbing Wang, Chenghu Zhou, "Threshold-based Routing-Topology Co-design for Optical Data Center," in IEEE/ACM Transactions on Networking, Vol. 31, No. 6, December, 2023. (Corresponding Author)
Min Yee Teh, Shizhen Zhao, Peirui Cao, Keren Bergman, "Enabling Quasi-static Reconfigurable Networks with Robust Topology Engineering," in IEEE/ACM Transactions on Networking, Vol. 31, No. 3, June, 2022. (Corresponding Author)

2027

Shengkai Lin, Kairui Zhou, Hongtao Zhang, Yibo Wu, Yi Pan, Yihan Yang, Qinwei Yang, Wei Zhang, Arvind Krishnamurthy, Shizhen Zhao, "SHIFT: Cross-NIC Fault Tolerance for RDMA Networks," in EuroSys, Rabat, Morocco, April, 2027. (Corresponding Author)
Ximeng Liu, Yingming Mao, Zhuoran Liu, Jinhui Ren, Shuyao Qi, Xinchi Han, Siyuan Feng, Peirui Cao, Yatao Li, Shizhen Zhao, Xinbing Wang, "AICCL: Pushing the Quality-Scalability-Flexibility Frontier in Collective Communication via Learning," in NSDI, Providence, RI, USA, May, 2027. (Corresponding Author)

2026

Ximeng Liu, Zhuoran Liu, Yingming Mao, Yatao Li, Shizhen Zhao, Xinbing Wang, "Geminet: Learning the Duality-based Topology-Agnostic Update Operator for Lightweight Traffic Engineering in Changing Topologies," in NSDI, Santa Clara, CA, USA, March, 2026. (Corresponding Author)
Yingming Mao, Qiaozhu Zhai, Ximeng Liu, Xinchi Han, Fanfan Li, Shizhen Zhao, Yuzhou Zhou, Zhen Yao, Xia Zhu, "ATRO: A Fast Algorithm for Topology Engineering of Reconfigurable Datacenter Networks," in INFOCOM, Tokyo, Japan, May, 2026.
Haonan Qi, Bin Lu, Yimian Hu, Ze Zhao, Meng Jin, Lixin Qu, Shuai Li, Shizhen Zhao, Lei Zhou, Xiaoying Gan, Xinbing Wang, Chenghu Zhou, "From Surface to Depth: Diagnosing Ocean Vertical Velocity with 3D Frequency Operator," in KDD, Jeju, Korea, August, 2026.
Haoyuan Liu, Kairui Zhou, Shuyao Qi, Qinwei Yang, Shengkai Lin, Shizhen Zhao, Wei Zhang, "LiveR: Fine-Grained Elasticity via Live Reconfiguration for Model Training," in arXiv preprint arXiv:2605.22014, May, 2026.
Likai Wang, Zhe Wang, Yimu Yuan, Siyu Yan, Shuhan Tian, Ning Liu, Hui Yuan, Linghe Kong, Qiao Xiang, Shizhen Zhao, Di Qu, Hexiang Song, Xiaolong Zheng, Yashar Ganjali, Guihai Chen, "Loss-Tolerant RDMA Network Over Commodity Devices," in IEEE/ACM Transactions on Networking, Vol. 34, 2026.
Cong Liang, Xiangli Song, Jing Cheng, Rui Xu, Mowei Wang, Yashe Liu, Zhenhua Liu, Shizhen Zhao, Yong Cui, "NegotiaToR: Toward a Simple Yet Effective On-Demand Reconfigurable Datacenter Network," in IEEE/ACM Transactions on Networking, Vol. 34, 2026.
Xinchi Han, Weihao Jiang, Yingming Mao, Yike Liu, Zhuoran Liu, Yongxi Lv, Peirui Cao, Zhuotao Liu, Ximeng Liu, Xinbing Wang, Changbo Wu, Zihan Zhu, Dongchao Wu, Jian Yang, Zhanbang Zhang, Yuansen Cheng, Shizhen Zhao, "Avoiding routing polarization for OCS-based GPU clusters," in Journal of Optical Communications and Networking, vol. 18, no. 9, pp. 900-914, 2026. (Technical Report)
Shan Gao, Duo Lin, Yimu Yuan, Zhe Wang, Mingyuan Zhang, Ruiping Zuo, Shizhen Zhao, Linghe Kong, Zhuotong Li, Shuyun Qian, Weiguo Chen, Liang Xu, "CollScope: Dynamic Training Semantic Alignment for Non-Intrusive Straggler Diagnosis," in ICNP, Tempe, Arizona, USA, October, 2026.
Qinwei Yang, Peirui Cao, Ruyi Zhang, Zhaochen Zhang, Shizhen Zhao, Dong Zhang, Bingheng Yan, Chen Tian, "STON: Scaling Torus-Based AI Training Clusters via Optical Circuit Switches," in IEEE JCC, Tokyo, Japan, July, 2026.

2025

Xinchi Han, Yongxi Lv, Weihao Jiang, Shuyuan Zhang, Yingming Mao, Shizhen Zhao, Zhuoran Liu, Zhuotao Liu, Peirui Cao, Ximeng Liu, Xinbing Wang, "A Highly Scalable LLM Clusters with Optical Interconnect," arXiv preprint arXiv:2411.01503, September, 2025.
Weihao Jiang, Wenli Xiao, Yuqing Yang, Peirui Cao, Shizhen Zhao, "Orderlock: A New Type of Deadlock and its Implications on High Performance Network Protocol Design," in SIGCOMM, Coimbra, Portugal, September, 2025. (Corresponding Author)
Shengkai Lin, Shizhen Zhao, Peirui Cao, Xinchi Han, Quan Tian, Wenfeng Liu, Qi Wu, Donghai Han, Xinbing Wang, "ONCache: A Cache-Based Low-Overhead Container Overlay Network," in NSDI, Philadelphia, PA, USA, April, 2025. (Corresponding Author)
Xiwei Wu, Yueyang Feng, Tianyi Huang, Xiaoyang Lu, Shengkai Lin, Lihan Xie, Shizhen Zhao, Qinxiang Cao, "VEP: A Two-stage Verification Toolchain for Full eBPF Programmability," in NSDI, Philadelphia, PA, USA, April, 2025. (Corresponding Author)
Shizhen Zhao, Ximeng Liu, Tianyu Zhu, Xinbing Wang, "Pulse+: DetNet Routing Under Delay-diff Constraint," in IEEE/ACM Transactions on Networking, Vol. 33, No. 6, December, 2025. (Technical Report)
Wenli Xiao, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "Segment EDF: A Scheduling Policy with Tight Deterministic Latency under Multi-hop Networks," in IEEE/ACM Transactions on Networking, Vol. 33, No. 1, February, 2025. (Corresponding Author)
Xinchi Han, Shizhen Zhao, Yongxi Lv, Peirui Cao, Weihao Jiang, Qinwei Yang, Yunzhuo Liu, Shengkai Lin, Bo Jiang, Ximeng Liu, Yong Cui, Chenghu Zhou, Yunzhuo Liu, Xinbing Wang, "vClos: Network Contention Aware Scheduling for Distributed Machine Learning Tasks in Multi-tenant GPU Clusters," in Computer Networks, Vol. 268, 2025. (Corresponding Author) (Technical Report)
Xiyuan Liu, Yang Liu, Jingyi Cheng, Ximeng Liu, Shizhen Zhao, "FauTE: Fault-tolerant Traffic Engineering in Data Center Network," in APNet, Shanghai, China, August, 2025. (Corresponding Author)

2024

Ximeng Liu, Shizhen Zhao, Yong Cui, Xinbing Wang, "FIGRET: Fine-Grained Robustness-Enhanced Traffic Engineering," in SIGCOMM, Sydney, NSW, Australia, August, 2024. (Corresponding Author)
Cong Liang, Xiangli Song, Jing Cheng, Mowei Wang, Yashe Liu, Zhenhua Liu, Shizhen Zhao, Yong Cui, "NegotiaToR: Towards A Simple Yet Effective On-demand Reconfigurable Datacenter Network," in SIGCOMM, Sydney, NSW, Australia, August, 2024.
Ximeng Liu, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "LPulse: An efficient algorithm for service function chain placement and routing with delay guarantee," in Computer Networks, Vol. 253, 2024.
Wenli Xiao, Yuqing Yang, Peirui Cao, Zhuoran Liu, Shizhen Zhao, Xinbing Wang, "OFC: An Original congestion-based Fine-grained Priority Flow Control," in SECON, Phoenix, AZ, USA, December, 2024. (Corresponding Author)
Shengkai Lin, Qinwei Yang, Zengyin Yang, Yuchuan Wang, Shizhen Zhao, "LubeRDMA: A Fail-safe Mechanism of RDMA," in APNet, Sydney, Australia, August, 2024. (Corresponding Author)
Peirui Cao, Wenxue Cheng, Shizhen Zhao, Yongqiang Xiong, "Network Load Balancing with Parallel Flowlets for AI Training Clusters," in SIGCOMM Workshop on Networks for AI Computing (NAIC), Sydney, Australia, August, 2024.
姚畅, 郝艳妮, 彭升辉, 牛志昂, 谢勇, 赵世振, "《信息系统架构发展展望——以国家自然科学基金委员会信息系统为例》," in 计算机科学, 2024. (Corresponding Author)

2023

Shizhen Zhao, Qizhou Zhang, Peirui Cao, Xiao Zhang, Xinbing Wang, Chenghu Zhou, "Flattened Clos: Designing High-performance Deadlock-free Expander Data Center Networks Using Graph Contraction," in NSDI, Boston, MA, USA, April, 2023.
Yunzhuo Liu, Bo Jiang, Shizhen Zhao, Tao Lin, Xinbing Wang, Chenghu Zhou, "LIBRA: Contention-Aware GPU Thread Allocation for Data Parallel Training in High Speed Networks," in INFOCOM, New York, NY, USA, May, 2023.
Peirui Cao, Shizhen Zhao, Dai Zhang, Zhuotao Liu, Mingwei Xu, Min Yee Teh, Yunzhuo Liu, Xinbing Wang, Chenghu Zhou, "Threshold-based Routing-Topology Co-design for Optical Data Center," in IEEE/ACM Transactions on Networking, Vol. 31, No. 6, December, 2023.
Min Yee Teh, Shizhen Zhao, Peirui Cao, Keren Bergman, "Enabling Quasi-static Reconfigurable Networks with Robust Topology Engineering," in IEEE/ACM Transactions on Networking, Vol. 31, No. 3, June, 2023.
Guangnan Feng, Dezun Dong, Shizhen Zhao, Yutong Lu, "GRAP: Group-level Resource Allocation Policy for Reconfigurable Dragonfly Network in HPC," in ICS, Orlando, FL, USA, June, 2023.
Shuyuan Zhang, Shu Shan, Shizhen Zhao, "Reducing Reconfiguration Time in Hybrid Optical-Electrical Datacenter Networks," in APNet, Hong Kong, China, June, 2023. (Corresponding Author)
Xiao Zhang, Peirui Cao, Yongxi Lyu, Qizhou Zhang, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "FC+: Near-optimal Deadlock-free Expander Data Center Networks," in ISPA, Wuhan, China, December, 2023. (Corresponding Author)
Tianyi Huang, Shizhen Zhao, "Accelerating QUIC with AF_XDP," in ICA3PP, Tianjin, China, October, 2023.

2022

Shizhen Zhao, Peirui Cao, Xinbing Wang, "Understanding the Performance Guarantee of Physical Topology Design for Optical Circuit Switched Data Centers," in SIGMETRICS, Mumbai, India, June, 2022.

2021

Peirui Cao, Shizhen Zhao, Min Yee Teh, Yunzhuo Liu, Xinbing Wang, "TROD: Evolving From Electrical Data Center to Optical Data Center," in ICNP, Virtual Event, November, 2021. (Corresponding Author)
Shizhen Zhao, Xiao Zhang, Peirui Cao, Xinbing Wang, "Design of Robust and Efficient Edge Server Placement and Server Scheduling Policies," in IWQoS, Virtual Event, June, 2021. (Technical Report)

2020

Shizhen Zhao, Yihan Zou, Xiaojun Lin, Dionysios Aliprantis, Hugo N. Villegas Pico, Minghua Chen, Anya Castillo, "Leveraging Generators With Complementary Capabilities for Robust Multistage Power Grid Operations," in IEEE Transactions on Control of Network Systems, Vol. 7, No. 3, September, 2020.
Min Yee Teh, Shizhen Zhao, Peirui Cao, Keren Bergman, "COUDER: Robust Topology Engineering for Optical Circuit Switched Data Center Networks," in arXiv preprint arXiv:2010.00090, October, 2020. (Corresponding Author)

2019

Shizhen Zhao, Rui Wang, Junlan Zhou, Joon Ong, Jeffery C. Mogul, Amin Vahdat, "Minimal Rewiring: Efficient Live Expansion for Clos Data Center Networks," in NSDI, Boston, MA, USA, February, 2019. (Technical Report)

2017

Shizhen Zhao, Xiaojun Lin, Minghua Chen, "Robust Online Algorithms for Peak-Minimizing EV Charging under Multi-Stage Uncertainty," in IEEE Transactions on Automatic Control, Vol. 62, No. 11, November, 2017. (Technical Report)
Lei Deng, Chih-Chun Wang, Minghua Chen, and Shizhen Zhao, "Timely Wireless Flows with General Traffic Patterns: Capacity Region and Scheduling Algorithms," in IEEE/ACM Transactions on Networking, Vol. 25, No. 6, December, 2017.

2013

Shizhen Zhao, Xiaojun Lin, Minghua Chen, "Peak-Minimizing Online EV Charging," in 51st Annual Allerton Conference on Communication, Control, and Computing, Monticello, IL, USA, October, 2013. (Technical Report)

2012

Shizhen Zhao, Xiaojun Lin, "On the Design of Scheduling Algorithms for End-to-End Backlog Minimization in Multi-hop Wireless Networks," in INFOCOM, Orlando, FL, USA, March, 2012. (Technical Report)
Yaokun Wu, Shizhen Zhao, "Incidence Matrix and Cover Matrix of Nested Interval Orders," in Electronic Journal of Linear Algebra, Vol. 25, pp. 90-104, 2012.

2011

Shizhen Zhao, Luoyi Fu, Xinbing Wang, Qian Zhang, "Fundamental Relationship between Node Density and Delay in Wireless Ad Hoc Networks with Unreliable Links," in MOBICOM, Las Vegas, NV, USA, September, 2011. (Technical Report)

想了解本人的研究方向，建议先阅读光电混合组网方面的介绍文章：OCS_review.pdf。以下按五个方向分类列出相关论文。

拓扑组网 / Topology Engineering

光电路交换（OCS）与光电混合数据中心网络架构、拓扑设计、重配置与扩展性研究。

Xinchi Han, Yongxi Lv, Weihao Jiang, Shuyuan Zhang, Yingming Mao, Shizhen Zhao, Zhuoran Liu, Zhuotao Liu, Peirui Cao, Ximeng Liu, Xinbing Wang, "A Highly Scalable LLM Clusters with Optical Interconnect," arXiv preprint arXiv:2411.01503, September, 2025.
Yingming Mao, Qiaozhu Zhai, Ximeng Liu, Xinchi Han, Fanfan Li, Shizhen Zhao, Yuzhou Zhou, Zhen Yao, Xia Zhu, "ATRO: A Fast Algorithm for Topology Engineering of Reconfigurable Datacenter Networks," in INFOCOM, Tokyo, Japan, May, 2026.
Peirui Cao, Shizhen Zhao, Dai Zhang, Zhuotao Liu, Mingwei Xu, Min Yee Teh, Yunzhuo Liu, Xinbing Wang, Chenghu Zhou, "Threshold-based Routing-Topology Co-design for Optical Data Center," in IEEE/ACM Transactions on Networking, Vol. 31, No. 6, December, 2023.
Min Yee Teh, Shizhen Zhao, Peirui Cao, Keren Bergman, "Enabling Quasi-static Reconfigurable Networks with Robust Topology Engineering," in IEEE/ACM Transactions on Networking, Vol. 31, No. 3, June, 2023.
Shizhen Zhao, Peirui Cao, Xinbing Wang, "Understanding the Performance Guarantee of Physical Topology Design for Optical Circuit Switched Data Centers," in SIGMETRICS, Mumbai, India, June, 2022.
Shizhen Zhao, Rui Wang, Junlan Zhou, Joon Ong, Jeffery C. Mogul, Amin Vahdat, "Minimal Rewiring: Efficient Live Expansion for Clos Data Center Networks," in NSDI, Boston, MA, USA, February, 2019. (Technical Report)
Shuyuan Zhang, Shu Shan, Shizhen Zhao, "Reducing Reconfiguration Time in Hybrid Optical-Electrical Datacenter Networks," in APNet, Hong Kong, China, June, 2023. (Corresponding Author)
Cong Liang, Xiangli Song, Jing Cheng, Mowei Wang, Yashe Liu, Zhenhua Liu, Shizhen Zhao, Yong Cui, "NegotiaToR: Towards A Simple Yet Effective On-demand Reconfigurable Datacenter Network," in SIGCOMM, Sydney, NSW, Australia, August, 2024.

路由与网络负载均衡 / Routing & Load Balancing

流量工程、路由算法、拥塞控制与数据中心网络负载均衡。

Ximeng Liu, Zhuoran Liu, Yingming Mao, Yatao Li, Shizhen Zhao, Xinbing Wang, "Geminet: Learning the Duality-based Topology-Agnostic Update Operator for Lightweight Traffic Engineering in Changing Topologies," in NSDI, Santa Clara, CA, USA, March, 2026. (Corresponding Author)
Xiyuan Liu, Yang Liu, Jingyi Cheng, Ximeng Liu, Shizhen Zhao, "FauTE: Fault-tolerant Traffic Engineering in Data Center Network," in APNet, Shanghai, China, August, 2025. (Corresponding Author)
Ximeng Liu, Shizhen Zhao, Yong Cui, Xinbing Wang, "FIGRET: Fine-Grained Robustness-Enhanced Traffic Engineering," in SIGCOMM, Sydney, NSW, Australia, August, 2024. (Corresponding Author)
Xinchi Han, Weihao Jiang, Yingming Mao, Yike Liu, Zhuoran Liu, Yongxi Lv, Peirui Cao, Zhuotao Liu, Ximeng Liu, Xinbing Wang, Changbo Wu, Zihan Zhu, Dongchao Wu, Jian Yang, Zhanbang Zhang, Yuansen Cheng, Shizhen Zhao, "Avoiding routing polarization for OCS-based GPU clusters," in Journal of Optical Communications and Networking, vol. 18, no. 9, pp. 900-914, 2026. (Technical Report)
Xinchi Han, Shizhen Zhao, Yongxi Lv, Peirui Cao, Weihao Jiang, Qinwei Yang, Yunzhuo Liu, Shengkai Lin, Bo Jiang, Ximeng Liu, Yong Cui, Chenghu Zhou, Yunzhuo Liu, Xinbing Wang, "vClos: Network Contention Aware Scheduling for Distributed Machine Learning Tasks in Multi-tenant GPU Clusters," in Computer Networks, Vol. 268, 2025. (Corresponding Author) (Technical Report)
Peirui Cao, Wenxue Cheng, Shizhen Zhao, Yongqiang Xiong, "Network Load Balancing with Parallel Flowlets for AI Training Clusters," in SIGCOMM Workshop on Networks for AI Computing (NAIC), Sydney, Australia, August, 2024.
Shizhen Zhao, Ximeng Liu, Tianyu Zhu, Xinbing Wang, "Pulse+: DetNet Routing Under Delay-diff Constraint," in IEEE/ACM Transactions on Networking, Vol. 33, No. 6, December, 2025. (Technical Report)
Wenli Xiao, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "Segment EDF: A Scheduling Policy with Tight Deterministic Latency under Multi-hop Networks," in IEEE/ACM Transactions on Networking, Vol. 33, No. 1, February, 2025. (Corresponding Author)
Ximeng Liu, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "LPulse: An efficient algorithm for service function chain placement and routing with delay guarantee," in Computer Networks, Vol. 253, 2024.

RDMA协议 / RDMA Protocols

RDMA可靠传输、无损网络协议、死锁避免与故障容错。

Shengkai Lin, Kairui Zhou, Hongtao Zhang, Yibo Wu, Yi Pan, Yihan Yang, Qinwei Yang, Wei Zhang, Arvind Krishnamurthy, Shizhen Zhao, "SHIFT: Cross-NIC Fault Tolerance for RDMA Networks," in EuroSys, Rabat, Morocco, April, 2027. (Corresponding Author)
Shengkai Lin, Qinwei Yang, Zengyin Yang, Yuchuan Wang, Shizhen Zhao, "LubeRDMA: A Fail-safe Mechanism of RDMA," in APNet, Sydney, Australia, August, 2024. (Corresponding Author)
Likai Wang, Zhe Wang, Yimu Yuan, Siyu Yan, Shuhan Tian, Ning Liu, Hui Yuan, Linghe Kong, Qiao Xiang, Shizhen Zhao, Di Qu, Hexiang Song, Xiaolong Zheng, Yashar Ganjali, Guihai Chen, "Loss-Tolerant RDMA Network Over Commodity Devices," in IEEE/ACM Transactions on Networking, Vol. 34, 2026.
Weihao Jiang, Wenli Xiao, Yuqing Yang, Peirui Cao, Shizhen Zhao, "Orderlock: A New Type of Deadlock and its Implications on High Performance Network Protocol Design," in SIGCOMM, Coimbra, Portugal, September, 2025. (Corresponding Author)
Shizhen Zhao, Qizhou Zhang, Peirui Cao, Xiao Zhang, Xinbing Wang, Chenghu Zhou, "Flattened Clos: Designing High-performance Deadlock-free Expander Data Center Networks Using Graph Contraction," in NSDI, Boston, MA, USA, April, 2023.
Xiao Zhang, Peirui Cao, Yongxi Lyu, Qizhou Zhang, Shizhen Zhao, Xinbing Wang, Chenghu Zhou, "FC+: Near-optimal Deadlock-free Expander Data Center Networks," in ISPA, Wuhan, China, December, 2023. (Corresponding Author)
Wenli Xiao, Yuqing Yang, Peirui Cao, Zhuoran Liu, Shizhen Zhao, Xinbing Wang, "OFC: An Original congestion-based Fine-grained Priority Flow Control," in SECON, Phoenix, AZ, USA, December, 2024. (Corresponding Author)

集合通信 / Collective Communication

分布式AI训练中的集合通信优化、通信库设计与训练性能诊断。

Ximeng Liu, Yingming Mao, Zhuoran Liu, Jinhui Ren, Shuyao Qi, Xinchi Han, Siyuan Feng, Peirui Cao, Yatao Li, Shizhen Zhao, Xinbing Wang, "AICCL: Pushing the Quality-Scalability-Flexibility Frontier in Collective Communication via Learning," in NSDI, Providence, RI, USA, May, 2027. (Corresponding Author)
Shan Gao, Duo Lin, Yimu Yuan, Zhe Wang, Mingyuan Zhang, Ruiping Zuo, Shizhen Zhao, Linghe Kong, Zhuotong Li, Shuyun Qian, Weiguo Chen, Liang Xu, "CollScope: Dynamic Training Semantic Alignment for Non-Intrusive Straggler Diagnosis," in ICNP, Tempe, Arizona, USA, October, 2026.
Yunzhuo Liu, Bo Jiang, Shizhen Zhao, Tao Lin, Xinbing Wang, Chenghu Zhou, "LIBRA: Contention-Aware GPU Thread Allocation for Data Parallel Training in High Speed Networks," in INFOCOM, New York, NY, USA, May, 2023.

eBPF

基于eBPF/AF_XDP的网络数据面可编程性、程序验证、协议加速与容器网络优化。

Shengkai Lin, Shizhen Zhao, Peirui Cao, Xinchi Han, Quan Tian, Wenfeng Liu, Qi Wu, Donghai Han, Xinbing Wang, "ONCache: A Cache-Based Low-Overhead Container Overlay Network," in NSDI, Philadelphia, PA, USA, April, 2025. (Corresponding Author)
Xiwei Wu, Yueyang Feng, Tianyi Huang, Xiaoyang Lu, Shengkai Lin, Lihan Xie, Shizhen Zhao, Qinxiang Cao, "VEP: A Two-stage Verification Toolchain for Full eBPF Programmability," in NSDI, Philadelphia, PA, USA, April, 2025. (Corresponding Author)
Tianyi Huang, Shizhen Zhao, "Accelerating QUIC with AF_XDP," in ICA3PP, Tianjin, China, October, 2023.