ACM SIGCOMM 2024 | CCF-A, acceptance rate 16.9% (62/366)
vPIFO: Virtualized Packet Scheduler for Programmable Hierarchical Scheduling in High-Speed Networks
Zhiyu Zhang, Shili Chen, Ruyi Yao, Ruoshi Sun, Hao Mei, Hao Wang, Zixuan Chen, Gaojian Fang, Yibo Fan, Wanxin Shi, Sen Liu, and Yang Xu
A hardware virtualization solution for programmable hierarchical packet scheduling in MTDCs.
APNet 2024 | CCF-C, acceptance rate 27.3% (24/88)
UniFL: Enabling Loss-tolerant Transmission in Federated Learning
Zixuan Chen, Yifan Ruan, Sen Liu, and Yang Xu
A loss-tolerant transmission protocol designed for Federated Learning.
ACM Eurosys 2024 | CCF-A, acceptance rate -% (-/-)
Halflife: An Adaptive Flowlet-based Load Balancer with Fading Timeout in DCNs
Sen Liu, Yongbo Gao, Zixuan Chen, Jiarui Ye, Haiyang Xu, Furong. Liang, Wei Yan, Zerui Tian, Quanwei Sun, Zehua Guo, Yang Xu
A novel flowlet-based load balancer that leverages fading FTV to reroute traffic promptly under different workloads without any prior knowledge.
IEEE IPDPS 2024 | CCF-B, acceptance rate -% (-/-)
MUSE: A Runtime Incrementally Reconfigurable Network Adapting to HPC Real-Time Traffic
Zijian Li, Zixuan Chen, Yiying Tang, Xin Ai, Yuanyi Zhu, Zhigao Zhao, Jiang Shao, Guowei Liu, Sen Liu, Bin Liu, and Yang Xu
A Dragonfly-based runtime incrementally reconfigurable network topology with optical circuit switch (OCS).
IEEE CLUSTER 2023 | CCF-B, acceptance rate 24.6% (32/130)
SDT: A Low-cost and Topology-reconfigurable Testbed for Network Research
Zixuan Chen, Zhigao Zhao, Zijian Li, Jiang Shao, Sen Liu, and Yang Xu
Software Defined Topology Testbed (SDT) offers a scalable, cost-effective, and time-efficient solution for conducting diverse network research experiments.
ACM TURC 2023 | Invited poster
Augmenting Distributed AI Training with Loss-tolerant Transmission
Zixuan Chen, Lei Shi, Yongbo Gao, Xuandong Liu, Xin Ai, Sen Liu, and Yang Xu
ACM ICPP 2023 | CCF-B, Best Paper candidate
OSP: Boosting Distributed Model Training with 2-stage Synchronization
Zixuan Chen, Lei Shi, Xuandong Liu, Jiahui Li, Sen Liu, and Yang Xu
A synchronization approach that enhances the throughput of PS-based DDL training by overlapping the communication with the computation by a 2-stage synchronization strategy.
IEEE/ACM IWQoS 2023 | CCF-B, acceptance rate 23.5% (62/264)
Boosting Distributed Machine Learning Training Through Loss-tolerant Transmission Protocol
Zixuan Chen, Lei Shi, Xuandong Liu, Xin Ai, Sen Liu, and Yang Xu
A customized transmission protocol that improves distributed machine learning throughput by properly exploiting link packet loss.
IEEE/ACM IWQoS 2023 | CCF-B, acceptance rate 23.5% (62/264)
Rusen: Rule Semantics Enabler toward Fast TCAM Update for Commodity SDN Switches
Hao Mei, Ruoshi Sun, Ruyi Yao, Chuhao Chen, Cong Luo, Zixuan Chen, Jiahui Li, Sen Liu, and Yang Xu
A rule semantics enabler enables multiple TCAM update algorithms in commodity SDN switches, which works as a transparent middle layer without modification to network devices.