NSDI '27 Spring Accepted Papers

NSDI '27 offers authors the choice of two submission deadlines. The list of accepted papers from the spring deadline is available below.

SkyVoyager: On using Multi-Region Spot Instances to Minimize AI Batch Job Cost

Zhifei Li, Renmin University of China and University of California, Berkeley; Tian Xia and Ziming Mao, University of California, Berkeley; Zihan Zhou, Shanghai Jiao Tong University; Ethan J. Jackson, Jamison Kerney, and Zhanghao Wu, University of California, Berkeley; Pratik Mishra, AMD; Yi Xu, Yifan Qiao, and Ion Stoica, University of California, Berkeley

A Decade Behind the Scenes: Evolving Layer-4 Load Balancers in Alibaba Cloud

Chengyuan Huang, Nanjing University; Rong Wen, Fudan University and Alibaba Group; Shuai Jin, Qingzheng Hou, Yahua Li, Kaiken Li, Shize Zhang, Enge Song, Bengbeng Xue, Donglin Lai, Jin Ke, Xiaoqing Sun, Jianyuan Lu, and Song Yang, Alibaba Group; Xing Li, Zhejiang University and Alibaba Group; Zhigang Zong, Biao Lyu, Shunmin Zhu, and Tian Pan, Alibaba Group; Xiaoliang Wang, Nanjing University; Dehui Wei, National University of Singapore; Qingkai Meng and Peirui Cao, Nanjing University; Keqiang He, Shanghai Jiao Tong University; Chen Tian and Guihai Chen, Nanjing University

Revisiting Network Support for Packet-level Load-balancing in RoCE

Jinchao Ma, Xi'an Jiaotong University; Ran Shu, Microsoft Research Asia; Danfeng Shan, Xi'an Jiaotong University; Wenxue Cheng, Microsoft Research Asia; Jinnuo Du, Xi'an Jiaotong University; Zhixiong Niu, Microsoft Research Asia; Peng Zhang, Xi'an Jiaotong University; Yongqiang Xiong, Microsoft Research Asia

R-Xray: Diagnosing Undiscovered Transport-Layer Anomalies in SmartNICs

Boliang Liu, Fudan University; Hanlin Huang and Peiwen Yu, Alibaba Group; Zhiyi Yao, Fudan University; Gang Lu, Alibaba Group; Yuedong Xu and Jun Wu, Fudan University; Jinlong Hou and Yuan Cheng, Shanghai Innovation Institute; Lijun Yu, Lei Wang, and John Cui, NVIDIA; Chang Liu, Feiyang Xue, and Binzhang Fu, Alibaba Group

Programmable Switch Telemetry at Microsecond Granularity: BurstMon in Action

Yinchuan Cong, Kun Xie, Jiajun He, Jingyang Lin, Xin Zeng, and Lele Wang, Hunan University; Jigang Wen, Hunan University of Science and Technology; Kenli Li, College of Information Science and Engineering, National Supercomputing Center in Changsha, Hunan University; Shiming He, Changsha University of Science and Technology; Wei Liang, Hunan University of Science and Technology; Gaogang Xie, Computer Network Information Center, Chinese Academy of Sciences; and University of Chinese Academy of Sciences

PipeTune: An Efficient Performance Tuning Framework for Host Datapaths

Xinyang Huang, Hong Kong University of Science and Technology; Zhuobin Huang, University of Electronic Science and Technology of China; Qijing Li, Hong Kong University of Science and Technology; Junxue Zhang, University of Science and Technology of China; Zhenghang Ren, Hong Kong University of Science and Technology; Mengyu Ma, Inspur; Han Tian, University of Science and Technology of China; Bowen Liu and Kai Chen, Hong Kong University of Science and Technology

Pyxis: Unlocking Multi-NIC Bandwidth for Incast Traffic in AI Clusters

Wenxue Li, Zihao Wang, Zhaoxiang Bao, Zhenghang Ren, Yijun Sun, and Bowen Liu, Hong Kong University of Science and Technology; Han Tian and Junxue Zhang, University of Science and Technology of China; Xiaolin Chen, Qiongtai Normal University; Kai Chen, Hong Kong University of Science and Technology

Radar: A dynamic and scalable system for heterogeneous data processing

Frank Sifei Luan and Ron Yifeng Wang, University of California, Berkeley; Yile Gu, University of Washington; Ziming Mao, Charlotte Lin, and Amog Kamsetty, University of California, Berkeley; Hao Chen, Cheng Su, Balaji Veeramani, Scott Lee, and SangBin Cho, Anyscale; Clark Zinzow, Together AI; Eric Liang, Databricks; Ion Stoica, University of California, Berkeley; Stephanie Wang, University of Washington

BRAIN: A Unified and Centralized Approach to Automated Outage Declaration at Azure Scale

Cong Chen and Youjiang Wu, Microsoft Azure; Chang Lou, University of Virginia; Peng Huang, University of Michigan; Yingnong Dang, Feng Gao, Maira Babang, Rajive Kumar, John Socha-Leialoha, Azad Naik, Lauren Zhang, Jonathan Chen, Xiangyang Cao, UV Yadav, Zhiyong Hu, Yi Chen, Aditya Mate, Jasmine Vaghei, Francisco Mandujano Reyes, Akshit Bhalla, Honghua Chang, Bing Hu, Meng Jin, Hrishi Kulkarni, Shane Hu, Justin Bang, Yueli Lu, Andy Stumpp, Souvik Debnath, Yusuf Tinwala, Piyali Jana, Jeffrey Sun, Amandeep Singh, Gabe Wishnie, Rigel Carlson, Sumer Sen, Matthew Hetrick, Salome Jacob, Jeff Davis, Jian Zhang, Bryan Alexander, and Hong Gao, Microsoft Azure; Si Qin, Qingwei Lin, and Dongmei Zhang, Microsoft; Mathew John, Zhangwei Xu, and John Sheehan, Microsoft Azure

LFS: Learning-Based Flexible CPU Scheduler

Yeonju Ro, The University of Texas at Austin; Vincent Zhang, The University of Texas at Austin and Roblox; Venkat Arun and Daehyeok Kim, The University of Texas at Austin; Jiaxin Lin, Cornell University; Christopher J. Rossbach, The University of Texas at Austin and Microsoft; Atlas Wang and Aditya Akella, The University of Texas at Austin

Enabling Multi-Agent Network Operations for Mission-Critical Infrastructure with NOK

Ziqiu Luo, Tsinghua University; Zhongxu Guan, Tsinghua Shenzhen International Graduate School, Tsinghua University; Li Chen, Zhongguancun Laboratory; Dan Li, Tsinghua University; Yong Jiang, Tsinghua Shenzhen International Graduate School; Haifeng Sun and Yukai Miao, Zhongguancun Laboratory; Shizhe Tan, China Mobile Communications Group Heilongjiang Co.,Ltd; Jiang Wang, China Mobile Communications Group Co.,Ltd; Hailong Song, China Mobile Jiutian Artificial Intelligence Technology (Beijing) Co., Ltd; Changqiu Sun, Jiayu Zhao, Yingqi Zhang, Ge Zhang, Changbin Du, Yu Wang, and Hongbo Jia, China Mobile Communications Group Heilongjiang Co.,Ltd

Enabling Efficient Datapath Pipelines for RDMA NICs

Yaozhen Li, Xuming Zhang, and Yulong Li, Tianjin University; Wenxin Li, Tianjin University and LingBoZhiXin Technology (Changzhou) Co., Ltd.; Jin Zhang, Xiancheng Meng, and Jinte Ying, Tianjin University; Sheng Chen, Tianjin University and LingBoZhiXin Technology (Changzhou) Co., Ltd.; Keqiu Li, Tianjin University

Exploring the Initial Performance of NB-IoT NTN over GEO: Measurement and Analysis

Xiong Wang, University of Science and Technology of China; Yumin Du, Southern University of Science and Technology; Bin Gu, unaffiliated; Zheng Lin, University of Hong Kong; Xiaoyang Li and Yi Gong, Southern University of Science and Technology; Wei Gong, University of Science and Technology of China; Linghe Kong, Shanghai Jiao Tong University

Towards A Predictable Software-Centric vSwitch Data Plane in Large-Scale Heterogeneous Cloud Networks

Heng Pan, Computer Network Information Center, Chinese Academy of Sciences; Shuo Li and Zhaolun Gu, Baidu (China) Co., Ltd; Lei Xu and Junjie Hao, Baidu Online Network Technology (Beijing) Co., Ltd.; Peng Wu, Feifei Liu, Zilin Wang, Haokun Wang, Linhu Li, Heng Guo, Weiping Pan, Yingming Mao, Yang Yang, Xiaoxiao Liu, Weili Zhang, Fusheng Sun, Ran Zheng, and Xianjun Meng, Baidu (China) Co., Ltd; Dongbiao He, Computer Network Information Center, Chinese Academy of Sciences; Gaogang Xie, CNIC CAS; UCAS, China

SkyRDMA: Fully Offloaded Cloud RDMA Virtualization

Yibo Huang, University of Michigan; Lizhou Gao and Liang Wang, Tencent; Ziqian Liu, The University of Hong Kong; Chengxuan Pei, The Chinese Universities of Hong Kong, Shenzhen; Yaowen Guo, Hongchen Ren, Xin Li, Junpeng Zhang, Chao Pei, Chongyang Hong, and Zhaoxun Zhou, Tencent; Shihan Lin, University of Michigan; Yiming Qiu, The University of Hong Kong; Yunming Xiao, The Chinese University of Hong Kong, Shenzhen; Congcong Miao, Tencent

AICCL: Pushing the Quality-Scalability-Flexibility Frontier in Collective Communication via Learning

Ximeng Liu, Shanghai Jiao Tong University; Yingming Mao, Xi'an Jiaotong University; Zhuoran Liu, Shanghai Jiao Tong University; Jinhui Ren, Tsinghua University; Shuyao Qi and Xinchi Han, Shanghai Jiao Tong University; Siyuan Feng, Shanghai Innovation Institute; Peirui Cao, Nanjing University; Yatao Li, Zhongguancun Academy and Zhonguancun Institute of Artificial Intelligence; Shizhen Zhao and Xinbing Wang, Shanghai Jiao Tong University

From ASIC to Fleet: Lessons from Building and Operating a Hyperscaler NIC

Prankur Gupta, Alexander Duyck, Jakub Kicinski, Joseph Provine, Neal Peacock, Prabhakaran Ganesan, Rajiv Krishnamurthy, Chen Liu, Akshay Viswakumar, Timothy Vitkin, Jie Meng, Beatriz Padilla Hernandez, Michael Edwards, Andrei Kozlov, Viren Nathan, Tianyi Cui, Joy Chaoyue Xiong, Raul Hormazabal, Mohsin Bashir, and Fred Feng, Meta; Nathan Walker, unaffiliated; Lavin Khandelwal, Matt Maia, Milo Piazza, Mohanraj Thillainayagam, Mahmoud Mehr, Amithash Prasad, and Lee Trager, Meta

Pythia: Learning-driven Optimization of Virtual Network Programming in Clouds

Luyao Luo, Soochow University; Hongli Xu and Gongming Zhao, University of Science and Technology of China; He Huang, Soochow University; Long Li, University of Science and Technology of China; Liguang Xie, Independent; Min Chen, Xuwei Yang, and Yongqiang Yang, Huawei; Zhuolong Yu, University of Science and Technology of China

AION: A Framework for Industrial System Function Virtualization

Nikolaos Mitsakis, Siemens AG, University of Kassel; Philippe Buschmann, Ben Mecklenburg, Matthias Eichholz, Nico Schöpf, Fuzail Hamid Banday, Yannic Breiting, and Carlos Guimarães, Siemens AG; Ang Chen, University of Michigan; Andreas Blenk, Siemens AG; Oliver Hohlfeld, University of Kassel

NOMADIX: Disaggregated Data Plane Architecture for Global Cellular Connectivity

Andrew E. Ferguson, The University of Edinburgh; Viktoria Vomhoff and Stefan Geißler, University of Wuerzburg; Marcelo Bagnulo Braun, University Carlos III of Madrid; Jesus Omaña Iglesias, Telefonica Research; Steffen Gebert, EMnify; Mahesh Marina, University of Edinburgh; Matteo Varvello, Nokia; Andra Lutu, Telefonica Research

CPX: An Edge-based Design for Cellular Roaming

Tenzin Samten Ukyab, University of California, Berkeley; Wesley Woo and Shaddi Hasan, Virginia Tech; Sylvia Ratnasamy, University of California, Berkeley; Scott Shenker, University of California, Berkeley and International Computer Science Institute (ICSI)

On the Role of LLMs in Designing the Architecture of Networked Systems

Pratyush Sahu, Georgia Institute of Technology; Rahul Bothra, University of Illinois at Urbana–Champaign and Google; Venkat Arun, The University of Texas at Austin; Brighten Godfrey, University of Illinois at Urbana–Champaign; Akshay Narayan, Brown University; Ahmed Saeed, Georgia Institute of Technology

Hold On Loosely: Fast, Fault-Tolerant, and Cost-Efficient Reinforcement Learning over Decentralized Network

Chaoyi Ruan, Geng Luo, and Xinyi Wan, National University of Singapore; Long Zhao and Qinghe Wang, Anhui University; Jiaan Zhu, University of Science and Technology of China; Duling Xu, Renmin University of China; Guanbin Xu, Independent Researcher; Dehui Wei and Xiang Liu, National University of Singapore; Cheng Li, University of Science and Technology of China (USTC); Haifeng Sun, National University of Singapore; Liang Luo, Independent Researcher; Congcong Miao and Jialin Li, National University of Singapore