Technical Program
APNET 2025 Program ScheduleSession Time
| Keynote Talk | 50 mins total |
| Regular Paper | 10 mins talk + 3 mins Q&A (13 mins total) |
| Academic Insight Talk | 30 mins total |
| Industry Spotlight Talk | 30 mins total |
August 7, 2025 (Day 1)
| Time | Event |
| 8:30 - 8:45 am | Opening Remarks General/PC Chairs |
| 8:45 - 9:35 am | Keynote 1: From Prediction to Proof: Rethinking AI for Systems and Networks [Slides] [Video] Prof. Matthew Caesar (UIUC, ACM SIGCOMM Chair) |
| 9:35 - 10:30 am | Session: Reinventing RDMA Session Chair: Chen Tian |
| Exposing RDMA NIC Resources for Software-Defined Scheduling [Slides] [Video] Yibo Huang (University of Michigan); Yiming Qiu (University of Michigan / UC Berkeley); Yunming Xiao, Archit Bhatnagar (University of Michigan); Sylvia Ratnasamy (UC Berkeley); Ang Chen (University of Michigan) |
|
| Cache-Aware I/O Rate Control for RDMA [Slides] [Video] Qijing Li, Xinyang Huang, Bowen Liu, Pengbo Li, Junxue Zhang, Kai Chen (Hong Kong University of Science and Technology) |
|
| Orderly Management of Packets in RDMA by Eunomia [Video] Sana Mahmood, Jinqi Lu, Soudeh Ghorbani (Johns Hopkins University) |
|
| UCM: Fast and Maintainable User-space RDMA Connection Setup [Slides] [Video] Huijun Shen (Hunan University); Jian Yang, Zelong Yue (ByteDance Inc.); Xingyu Guo, Xijin Yin (Hunan University); Lang An, Yulin Chen, Jie Ding, Hongyu Wu, Yong Zhang, Jianxi Ye (ByteDance Inc.); Guo Chen (Hunan University) |
|
| 10:30 - 11:00 am | Coffee Break |
| 11:00 - 11:55 am | Session: Load, Balance, Repeat Session Chair: Henry Xu |
| Maat: A Fair Layer-4 Load Balancer With Per-Connection Consistency [Slides] [Video] Ju Huang, Lu Tang (Xiamen University) |
|
| Remote TCP Connection Offload with XO [Video] Shuo Li, Steven Chien, Tianyi Gao, Michio Honda (University of Edinburgh) |
|
| SRC: A Scalable Reliable Connection for RDMA with Decoupled QPs and Connections [Slides] [Video] Yiren Zhao, Ran Shu, Yongqiang Xiong (Microsoft Research) |
|
| Enabling Packet Spraying over Commodity RNICs with In-Network Support [Slides] [Video] Xiangzhou Liu, Wenxue Li, Kai Chen (Hong Kong University of Science and Technology) |
|
| 11:55 am - 1:15 pm | Lunch |
| 1:15 - 2:10 pm | Session: Smarter Control, Better Flow Session Chair: Alessandro Cornacchia |
| Congestion Control for AI Workloads with Message-Level Signaling [Slides] [Video] Yuxuan Li, Zhenghang Ren, Wenxue Li, Xiangzhou Liu, Kai Chen (Hong Kong University of Science and Technology) |
|
| Collaborative Multi-Flow Congestion Control via Deep Reinforcement Learning [Video] Qiangqiang Wei, Jiangping Han, Kaiping Xue, Jinhao Liu (University of Science and Technology of China); Naiqiang qiao, Hao Chen (HUAWEI TECHNOLOGIES CO., LTD.) |
|
| ORC: Online Reinforcement Learning for Congestion Control with Fast Convergence [Slides] [Video] Yijun Li, Jiawei Huang, Chuliang Wu (Central South university); Xiaojun Zhu, Jianxin Wang (Central South University) |
|
| Hybrid Congestion Control with Variable Monitoring Time Period [Slides] [Video] Xiaohong Qiu, Haifeng Liu (Jiangxi University of Science and Technology); Shuangmei Liu (Nanchang Hangkong University); Zhen Li, Xiaojun Zhu (Jiangxi University of Science and Technology) |
|
| 2:10 - 3:05 pm | Session: Networks in the Wild Session Chair: Shuai Wang |
| Assessing the Impact of ISP de-peering: A case study of Cogent's Disconnection from Russian Networks in Routing Perspective [Slides] [Video] Liming Liu (Tsinghua University, Zhongguancun Laboratory); Yuanyuan Zhang, Kun Guo, Meijia Hou (Zhongguancun Laboratory); Mingwei Xu (Tsinghua University, Zhongguancun Laboratory); Jiahao Cao (Tsinghua University); Xin Gao, Jiang Li, Yonghong Fu (Zhongguancun Laboratory) |
|
| Topology-Adaptive LEO Satellite Network Telemetry via Graph Isomorphism and Topology Partitioning [Slides] [Video] Yan Zhang (Purple Mountain Laboratories); Tian Pan (Beijing University of Posts and Telecommunications); Yan Zheng (Purple Mountain Laboratories); Guohao Ruan, Haonan Li (Beijing University of Posts and Telecommunications); Yi Liu (SICE, Beijing University of Posts and Telecommunications); Jiang Liu, Tao Huang (Beijing University of Posts and Telecommunications) |
|
| CTSVN: A Solution for Computation Task Scheduling in Vehicle Networking [Video] Zheng Fang, Kun Xie, Xiaohong Huang, Pei Zhang, Dandan Li (BUPT) |
|
| Understanding the Long Tail Latency of TCP in Large-Scale Cloud Networks [Slides] [Video] Zihao Fan (Shanghai Jiao Tong University and Alibaba Cloud); Enge Song (Alibaba Cloud); Bo Jiang (Shanghai Jiao Tong University); Yang Song, Yuke Hong, Bowen Yang, Yilong Lv, Yinian Zhou, Junnan Cai, Chao Wang, Yi Wang, Yehao Feng, Dian Fan, Ye Yang, Shize Zhang, Xiaoqing Sun, Jianyuan Lu (Alibaba Cloud); Xing Li (Zhejiang University and Alibaba Cloud); Jun Liang (Alibaba Cloud); Biao Lyu (Zhejiang University and Alibaba Cloud); Zhigang Zong (Alibaba Cloud); Shunmin Zhu (Hangzhou Feitian Cloud and Alibaba Cloud) |
|
| 3:05 - 4:00 pm | Poster Session + Coffee Break |
| 4:00 - 6:00 pm | Academic Insight Talks Session Chair: Kai Chen |
| Making Serverless Computing Efficient and Scalable [Slides] [Video] Prof. K. K. Ramakrishnan (Distinguished Professor, University of California, Riverside) |
|
| Congestion Management in/beyond Datacenter Networks: Pyrrha as a case [Slides] [Video] Prof. Chen Tian (Professor, Nanjing University) |
|
| What does sparsity mean for machine learning systems? Prof. Hong Xu (Associate Professor, The Chinese University of Hong Kong) |
|
| On the Design of High Performance Routing Protocols in AI Clusters [Slides] [Video] Prof. Shizhen Zhao (Associate Professor, Shanghai Jiao Tong University) |
|
| 6:00 - 6:15 pm | Break |
| 6:15 - 9:30 pm | APNet’25 Banquet |
| At 6:15 pm, depart from Crowne Plaza Hotel to Peace Hotel by shuttle bus; enjoy the banquet at Peace Hotel from 7:00 to 9:30 pm; and return to Crowne Plaza Hotel by shuttle bus at 9:30 pm. |
August 8, 2025 (Day 2)
| Time | Event |
| 8:30 - 9:20 am | Keynote 2: Democratizing Deep Learning Training: Towards Training LLMs using Consumer-grade GPUs [Video] Prof. Dongsu Han (KAIST) |
| 9:20 - 10:00 am | Session: Simulators for the AI Age Session Chair: Kaihui Gao |
| Miniature: Fast AI Supercomputer Networks Simulation on FPGAs [Slides] [Video] Yicheng Qian (Northeastern University/Microsoft Research); Ran Shu, Rui Ma, Yang Wang (Microsoft Research); Derek Chiou (UT Austin/Microsoft); Nadeen Gebara, Luca Piccolboni (Microsoft); Miriam Leeser (Northeastern University); Yongqiang Xiong (Microsoft Research) |
|
| Nüwa: Efficient Generative Control Plane for AI Network Simulation [Slides] [Video] Wenkai Li (Xi'an Jiaotong University); Ran Shu (Microsoft Research); Peng Zhang (Xi'an Jiaotong University); Yongqiang Xiong (Microsoft Research) |
|
| SplitNN: Single-Machine Network Emulation at Scale with Minute-Level Construction of 10K-Node Virtual Networks [Slides] [Video] Kaifei Peng, Yanbiao Li, Wenbin Li, Xian Yu, Gaogang Xie (Computer Network Information Center, Chinese Academy of Sciences; University of Chinese Academy of Sciences) |
|
| 10:00 - 10:40 am | Session: Redefining the Network Stack Session Chair: Yunming Xiao |
| OmniDMA: Scalable RDMA Transport over WAN [Slides] [Video] Kai Lv (Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences; Huawei); Jinyang Li (Huawei); Pengyi Zhang (Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences); Heng Pan (Computer Network Information Center, Chinese Academy of Sciences); Luyang Li (Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences); Shuihai Hu (Huawei); Zhenyu Li (Institute of Computing Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences.); Gaogang Xie (Computer Network Information Center, Chinese Academy of Sciences; University of Chinese Academy of Sciences); Jingbin Zhou, Kun Tan (Huawei) |
|
| Designing Transport-Level Encryption for Datacenter Networks [Slides] [Video] Tianyi Gao, Xinshu Ma, Suhas Narreddy, Eugenio Luo, Steven Chien, Michio Honda (University of Edinburgh) |
|
| Enhancing Network Traffic Prediction by Integrating Graph Transformer with a Temporal Model [Video] Xiucheng Sun, Runqun Xiong, Dian Shen (Southeast University); Junzhou Luo (Southeast University, Nanjing, P.R. China) |
|
| 10:40 - 11:10 am | Coffee Break |
| 11:10 am - 12:05 pm | Session: Efficient AI at the Edge and Core Session Chair: Li Chen |
| Mnemosyne: Lightweight and Fast Error Recovery for LLM Training in a Just-In-Time Manner [Slides] [Video] Jinyi Xia, Menghao Zhang, Jiaxun Huang, Yuezheng Liu (Beihang University); Xiaohe Hu (Infrawaves); Xudong Liu, Chunming Hu (Beihang University) |
|
| Rethinking Dynamic Networks and Heterogeneous Computing in Automatic Parallelization [Slides] [Video] Ruilong Wu, Xinjiao Li, Yisu Wang, Xinyu Chen, Dirk Kutscher (The Hong Kong University of Science and Technology (Guangzhou)) |
|
| FlexSpark: Robust and Efficient Multi-Device Collaborative Inference over Wireless Network [Video] Yiyang Shao, Hongyi Li, Shuihai Hu, Xinle Du, Hao Wu, Jingbin Zhou, Kun Tan (Huawei) |
|
| PromptMobile: Efficient Promptus for Low Bandwidth Mobile Video Streaming [Video] Liming Liu, Jiangkai Wu, Haoyang Wang, Peiheng Wang, Zongming Guo, Xinggong Zhang (Peking University) |
|
| 12:05 - 1:25 pm | Lunch |
| 1:25 - 2:20 pm | Session: Programmability and Performance Session Chair: Jiaqi Gao |
| Augmenting Public Cloud Infrastructure for Heterogeneous Network Function Virtualization [Slides] [Video] Haonan Li, Yang Song, Tian Pan, Zhigang Zong, Bengbeng Xue, Xionglie Wei, Yisong Qiao, Donglin Lai, Baohai Hu, Jin Ke, Enge Song, Yuxiang Lin, Xiaomin Wu, Jianyuan Lu (Alibaba Cloud); Xing Li, Biao Lyu (Zhejiang University and Alibaba Cloud); Rong Wen (Alibaba Cloud); Jiao Zhang, Tao Huang (Purple Mountain Laboratories); Shunmin Zhu (Hangzhou Feitian Cloud and Alibaba Cloud) |
|
| Cost-Efficient and Reliable SFC Orchestration in Mobile Edge Computing [Slides] [Video] Yuanfei Xiao, Zhenli He, Qixin Peng (Yunnan University) |
|
| Toward Scalable Learning-Based Optical Restoration [Slides] [Video] Siyong Huang (Xiamen University); Qingyu Song (The Chinese University of Hong Kong); Kexin Yu (Xiamen University); Zhaoning Wang (China Unicom); Zhizhen Zhong (Massachusetts Institute of Technology); Qiao Xiang, Jiwu Shu (Xiamen University) |
|
| Unmasking Vulnerabilities of HyperLogLog: Security via Parameter Extraction [Slides] [Video] Shishi Zhang, Ning Wang, Lu Tang (Xiamen University) |
|
| 2:20 - 3:15 pm | Session: Resilient Systems, Secure Designs Session Chair: Qiao Xiang |
| FauTE: Fault-tolerant Traffic Engineering in Data Center Network [Video] Xiyuan Liu (Nanyang Technological University); Yang Liu (Shanghai Institute of Satellite Engineering); Jingyi Cheng, Ximeng Liu, Shizhen Zhao (Shanghai Jiao Tong University) |
|
| A Theoretical Framework for Quantitative Evaluation of Padding Defenses against Website Fingerprinting [Video] Yue Gu (Tsinghua University/Tsinghua Shenzhen International Graduate School); Hao Wu (CNCERT/CC); Dan Li (Tsinghua University) |
|
| H-NRF: A High-performance and Evolutive NRF Framework for Large-scale Mobile Core Network [Video] Zhuoran Ma (Hunan University; CNIC CAS); Yanbiao Li (CNIC CAS; UCAS, China); Xin Wang (SUNY Stony Brook); Xian Yu, Xinyi Zhang, Shiyi Liu (CNIC CAS; UCAS China); Kun Xie (Hunan university); Gaogang Xie (CNIC CAS; UCAS, China) |
|
| Cloud Overbooking Optimization: Reducing Temporal Volatility through Spatial Workload Aggregation [Slides] [Video] Baoqing Wang, Long Li, Jing Yang, Jiawei Liu, Gongming Zhao, Hongli Xu (University of Science and Technology of China) |
|
| 3:15 - 3:45 pm | Coffee Break |
| 3:45 - 5:45 pm | Industry Spotlight Talks Session Chair: Kun Tan |
| Expressways: Unlocking the Full Potential of AI With Next-Gen Networking Dr. Gil Bloch (Senior Principal Architect at NVIDIA) |
|
| Compute Scaling for AI [Slides] [Video] Dr. Fan Yang (Senior Principal Research Manager at MSR-Asia) |
|
| Technical Challenges for AI Cluster Network [Slides] [Video] Dr. Binyang Liu (Director of Network Technology Lab, Huawei Central Research Institute) |
|
| Luoshen: A Large Scale High Performance Cloud Network for Public Cloud [Video] Dr. Shunmin Zhu (Head of Cloud Network Division at Alibaba Cloud) |
|
| 5:45 - 6:25 pm | Panel Discussion: Joint Academia & Industry |
| 6:25 - 6:30 pm | Concluding Remarks General/PC Chairs |
