Keqiang He

Keqiang He

Auto-generated profile, based on publicly available information

Keqiang He

Shanghai Jiao Tong University, Shanghai, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 August 2025
Verified
Conference Paper
Open access
CQChenchen Qi
Chenchen Qi
WWWenfei Wu
Wenfei Wu
YWYongcan Wang
Yongcan Wang
KHKeqiang He
Keqiang He
YKYu-Hsiang (Sean) Kao
Yu-Hsiang (Sean) Kao
SDShan Ding
Shan Ding

30 March 2025
Verified
Conference Paper
Open access
ZZZhaochen Zhang
Zhaochen Zhang
FXFeiyang Xue
Feiyang Xue
KHKeqiang He
Keqiang He
ZYZhimeng Yin
Zhimeng Yin
GAGianni Antichi
Gianni Antichi
CTChen Tian
Chen Tian

30 March 2025
Verified
Conference Paper
Open access
YCYanqing Chen
Yanqing Chen
LWLi Wang
Li Wang
JWJingzhi Wang
Jingzhi Wang
SLSongyue Liu
Songyue Liu
KHKeqiang He
Keqiang He
CTChen Tian
Chen Tian

4 February 2025 in Computer Networks
Verified
Journal Article
Closed access
MWMingyao Wang
Mingyao Wang
KHKeqiang He
Keqiang He
PCPeirui Cao
Peirui Cao
JDJiong Duan
Jiong Duan
DLDongliang Lv
Dongliang Lv
CTChen Tian
Chen Tian

29 December 2025 in IEEE Transactions on Networking
Verified
Journal Article
Closed access
CHChengyuan Huang
Chengyuan Huang
ZCZhengqi Cui
Zhengqi Cui
JXJun Xu
Jun Xu
ZZZhaochen Zhang
Zhaochen Zhang
LWLi Wang
Li Wang
CTChen Tian
Chen Tian
Distributed AI model training generates bursty, low-entropy elephant flows that challenge existing single-path transport protocols in multi-stage Clos networks, leading to congestion and inefficiency. Multipath transport emerges as a promising solution, leveraging multiple paths to balance traffic and enhance resilience. However, current multipath RDMA solutions suffer from scalability, congestion control, and load-balancing inefficiencies. This paper introduces Unified Delay-driven Multipath Protocol (UDMP), a novel approach that co-designs congestion control and load balancing using network delay as a unified signal. UDMP employs delay-gradient-based congestion control to precisely resolve unavoidable congestion. Moreover, UDMP leverages delay-assisted load balancing to shift traffic across paths with minimal latency adaptively, maintaining throughput when encountering avoidable congestion. A novel Token Pool design integrates these components, eliminating per-path state overhead while achieving fine-grained traffic distribution. Implementations on DPDK and NS3 demonstrate that UDMP achieves up to 2x higher throughput and reduces flow completion times by up to 30% compared to state-of-the-art methods like MPRDMA and QP-Scaling. These results highlight UDMP’s effectiveness in meeting the stringent performance requirements of modern distributed AI training workloads.

27 February 2026 in IEEE Transactions on Networking
Verified
Journal Article
Closed access
BCBrian Chang
Brian Chang
KHKeqiang He
Keqiang He
SCShawn Shuoshuo Chen
Shawn Shuoshuo Chen
JLJiaxin Lin
Jiaxin Lin
MZMingyang Zhang
Mingyang Zhang
AAAditya Akella
Aditya Akella
Many proposals have demonstrated the efficiency advantages of software-defined networking (SDN) in managing data center networks. Common practices employ centralized traffic engineering (TE) in the SDN control plane to optimize load balancing and throughput. Meanwhile, for high availability purposes, the control plane is partitioned to ensure the impact of a single faulty controller is contained. However, the interaction between these two aspects is often overlooked. In particular, we show that the current control plane partitioning approach leads to imbalanced link loads and degraded application performance. To address this issue, we propose virtual slicing, a new control plane partitioning scheme. Virtual slicing achieves desirable traffic engineering performance while retaining the availability guarantees from the current approach. Virtual slicing is implemented and evaluated with real-world and synthetic traffic traces on production spine-free data center networks. Results show that virtual slicing reduces tail link utilizations by up to 28.4%, and improves flow completion times by up to 36%.