Dong-Yang Yu 俞东阳
2nd PhD Student
Research interests include datacenter networks, and RDMA transmission optimization.

Hi! I am currently a second-year Ph.D. student at BUPT, where I am lucky to be advised by Prof. Yuchao Zhang. Before that, I earned my M.Eng degree in Computer Technology from Soochow University, where I was grateful to be supervised by Prof. Jin Wang. Prior to that, I received double B.S. degrees in Computer Science from Northern Arizona University and Computer Science & Technology from Yangzhou University in 2020, respectively.

I would love to work with all people together on some interesting projects! Feel free to drop me an email if you have any ideas to discuss!


News (view all )
2026
Apr. 2026, 🎉🎉🎉 My research is supported by the BUPT Excellent Ph.D. Students Foundation (5 awards out of 28 applicants).
Highlight
Mar. 2026, I am honored to receive the EuroSys 2026 Travel Grant! Looking forward to presenting our LCMP and meeting researchers in Edinburgh, UK this April.
Highlight
Feb. 2026, Thrilled to be selected for the inaugural IETF Chinese Youth Talent Program, co-sponsored by the Internet Society of China (ISC), Huawei, and others. I will be attending the IETF 125 meeting in Shenzhen to explore global Internet standards. [Link]
Highlight
Jan. 2026, Our paper "LCMP: Distributed Long-Haul Cost-Aware Multi-Path Routing for Inter-Datacenter RDMA Networks" is accepted by the 21th European Conference on Computer Systems (EuroSys, CCF-A) [Link]
Highlight
Selected Publications (view all )
LCMP: Distributed Long-Haul Cost-Aware Multi-Path Routing for Inter-Datacenter RDMA Networks
LCMP: Distributed Long-Haul Cost-Aware Multi-Path Routing for Inter-Datacenter RDMA Networks

Dong-Yang Yu, Yuchao Zhang*, Xiaodi Wang, Jun Wang, Wenfei Wu, Haipeng Yao, Wendong Wang, Ke Xu (* corresponding author)

21st European Conference on Computer Systems (EuroSys 2026), CCF-A (Acceptance Ratio 138/792=17.4%)

RDMA-empowered cloud services are gradually deployed across datacenters (DCs) with multiple paths, which exhibit new properties of path asymmetry, delayed congestion signals, and simultaneous flow routing collisions, and further fail existing routing methods. We present LCMP, a distributed long-haul cost-aware multi-path routing framework that aims to place RDMA flows on multiple inter-DC paths, achieving low-cost, low-latency, and congestion-responsive transmission. LCMP combines a control-plane path-quality score with compact on-switch congestion signals, where the former unifies quality assessment for asymmetric paths and the latter enables responsive reaction to path congestion. LCMP further resolves the simultaneous flow decision collision problem by filtering high-cost candidates, and performing a diversity-preserving hash inside the reduced set. On an 8-DC testbed, LCMP reduces median and tail FCT slowdown by up to 76% and 64%, respectively compared to state-of-the-art (SOTA) DCN routing strategies. And large-scale NS-3 simulations under the 2000 km inter-DC scenario confirm similar improvements.

LCMP: Distributed Long-Haul Cost-Aware Multi-Path Routing for Inter-Datacenter RDMA Networks

Dong-Yang Yu, Yuchao Zhang*, Xiaodi Wang, Jun Wang, Wenfei Wu, Haipeng Yao, Wendong Wang, Ke Xu (* corresponding author)

21st European Conference on Computer Systems (EuroSys 2026), CCF-A (Acceptance Ratio 138/792=17.4%)

RDMA-empowered cloud services are gradually deployed across datacenters (DCs) with multiple paths, which exhibit new properties of path asymmetry, delayed congestion signals, and simultaneous flow routing collisions, and further fail existing routing methods. We present LCMP, a distributed long-haul cost-aware multi-path routing framework that aims to place RDMA flows on multiple inter-DC paths, achieving low-cost, low-latency, and congestion-responsive transmission. LCMP combines a control-plane path-quality score with compact on-switch congestion signals, where the former unifies quality assessment for asymmetric paths and the latter enables responsive reaction to path congestion. LCMP further resolves the simultaneous flow decision collision problem by filtering high-cost candidates, and performing a diversity-preserving hash inside the reduced set. On an 8-DC testbed, LCMP reduces median and tail FCT slowdown by up to 76% and 64%, respectively compared to state-of-the-art (SOTA) DCN routing strategies. And large-scale NS-3 simulations under the 2000 km inter-DC scenario confirm similar improvements.

Education
  • Beijing University Of Posts and Telecommunications (BUPT)
    Beijing University Of Posts and Telecommunications (BUPT)
    State Key Laboratory of Networking and Switching Technology
    School of Computer Science (National Pilot Software Engineering School)
    Supervisor: Prof. Yuchao Zhang
    Ph.D. Student (Software Engineering)
    Sep. 2024 - Present
  • Soochow University
    Soochow University
    School of Computer Science & Technology
    Supervisor: Prof. Jin Wang
    M.Eng (Computer Technology)
    Sep. 2021 - Jun. 2024
  • Northern Arizona University
    Northern Arizona University
    School of Informatics, Computing, and Cyber Systems
    B.S. (Computer Science)
    Sep. 2018 - Jun. 2020
  • Yangzhou University
    Yangzhou University
    College of Information and Artificial Intelligence
    B.S. (Computer Science & Technology)
    Sep. 2017 - Jun. 2021
Honors & Awards
  • EuroSys 2026 Travel Grant
    Mar. 2026
  • IETF Chinese Youth Talent Program (Only 60 recipients in China)
    Mar. 2026