OSDI25 论文分类总结

OSDI25 论文分类总结

会议概览

OSDI 2025会议汇聚了系统软件领域的最新研究成果,共接收论文54篇,展现了该会议在操作系统与分布式系统领域的高度影响力与学术权威性。会议涵盖了多个关键技术方向,体现了系统研究的广泛覆盖面与深度交叉。 技术趋势上,AI与系统的融合成为一大亮点,AI技术正被广泛应用于系统优化与管理。分布式系统与数据中心、存储系统、操作系统内核等传统领域仍保持强劲活力,同时与调度管理、数据库系统紧密结合,推动高效能计算的发展。隐私与安全方向的研究持续升温,反映出对数据合规与系统防护的高度重视。研究热点包括智能系统设计、资源弹性调度、安全机制增强及绿色数据中心等,展现出系统领域对现实挑战的积极响应与持续创新。

数据生成时间:2025年08月31日

54
论文总数
8
技术领域
15
参与国家
72
主要机构

🔬 技术领域分类

技术领域分布

技术领域统计

🌍 作者来源分析

国家分布

主要机构

🔍 关键词分析

技术关键词热度

论文列表

AI + Systems (13篇)

KPerfIR: Towards a Open and Compiler-centric Ecosystem for GPU Kernel Performance Tooling on Modern AI Workloads

Authors: Yue Guan, University of California, San Diego; Yuanwei Fang, Meta; Keren Zhou, George Mason University and OpenAI; Corbin Robeck, AMD; Manman Ren, Meta; Zhongkai Yu, University of California, San Diego; Yufei Ding, University of California San Diego, Meta; Adnan Aziz, Meta

MISO: A Multi-Level Superoptimizer for Tensor Programs

Authors: Mengdi Wu and Xinhao Cheng, Carnegie Mellon University; Shengyu Liu and Chunan Shi, Peking University; Jianan Ji and Kit Ao, Carnegie Mellon University; Praveen Velliengiri, Pennsylvania State University; Xupeng Miao, Purdue University; Oded Padon, Weizmann Institute of Science; Zhihao Jia, Carnegie Mellon University

Write Once, Run Anywhere: Transcompiling Tensor Programs for Deep Learning Systems with a Neural-Symbolic Approach

Authors: Shouyang Dong, University of Science and Technology of China; Yuanbo Wen, Jun Bi, Di Huang and Jiaming Guo, Institute of Computing Technology, Chinese Academy of Sciences; Jianxing Xu and Ruibai Xu, University of Science and Technology of China; Xinkai Song, Yifan Hao, and Ling Li, Institute of Software, Chinese Academy of Sciences; Xuehai Zhou, University of Science and Technology of China; Tianshi Chen, Cambricon Technologies; Qi Guo and Yunji Chen, Institute of Computing Technology, Chinese Academy of Sciences;

LLMoC: Large Language Model Inference at Wafer Scale

Authors: Congjie He, Yeqi Huang, and Pei Mu, University of Edinburgh; Ziming Miao, Jilong Xue, Lingxiao Ma, and Fan Yang, Microsoft Research; Luo Mai, University of Edinburgh

Fast and Live Model Auto Scaling without Caching

Authors: Dingyan Zhang, Haotian Wang, Yang Liu, and Xingda Wei, Shanghai Jiao Tong University; Yizhou Shan, Huawei Cloud; Rong Chen and Haibo Chen, Shanghai Jiao Tong University

Bayesian Code Diffusion for Efficient Automatic Deep Learning Program Optimization

Authors: Isu Jeong and Seulki Lee, Ulsan National Institute of Science and Technology

Training with Confidence: Catching Silent DL Training Bugs with Automated Proactive Checks

Authors: Yuxuan Jiang, Ziming Zhou, Boyu Xu, Beijie Liu, Runhui Xu, and Peng Huang, University of Michigan

Neutrino: Fine-grained GPU Kernel Profiling via Programmable Probing

Authors: Songlin Huang and Chenshu Wu, The University of Hong Kong

Principles and Methodologies for System Performance Optimization

Authors: Sujin Park, Mingyu Guan, Xiang Cheng, and Taesoo Kim, Georgia Institute of Technology

Clover: Exploiting Intra-device Parallelism for High Throughput Large Language Model Serving

Authors: Kan Zhu, University of Washington; Yufei Gao, Tsinghua University and University of Washington; Yilong Zhao, University of California, Berkeley; Liangyu Zhao, University of Washington; Gefei Zuo, University of Michigan; Yile Gu and Dedong Xie, University of Washington; Tian Tang and Qinyu Xu, Tsinghua University and University of Washington; Zihao Ye, Keisuke Kamahori, and Chien-Yu Lin, University of Washington; Ziren Wang, Tsinghua University and University of Washington; Stephanie Wang, Arvind Krishnamurthy, and Baris Kasikci, University of Washington

PipeThreader: Software-Defined Pipelining for Efficient DNN Execution

Authors: Yu Cheng and Lei Wang, Peking University and Microsoft Research; Yining Shi, Peking University; Yuqing Xia, Lingxiao Ma, Jilong Xue, and Yang Wang, Microsoft Research; Zhiwen Mo, Imperial College London and Microsoft Research Asia; Feiyang Chen, Shanghai Jiao Tong University and Microsoft Research Asia; Fan Yang and Mao Yang, Microsoft Research; Zhi Yang, Peking University

WLB-LLM: Workload-Balanced 4D Parallelism for Large Language Model Training

Authors: Zheng Wang, University of California, San Diego; Anna Cai and Xinfeng Xie, Meta; Zaifeng Pan and Yue Guan, University of California, San Diego; Weiwei Chu, Jie Wang, Shikai Li, Jianyu Huang, Chris Cai, and Yuchen Hao, Meta; Yufei Ding, University of California, San Diego

DecDEC: A Systems Approach to Advancing Low-Bit LLM Quantization

Authors: Yeonhong Park, Jake Hyun, Hojoon Kim, and Jae W. Lee, Seoul National University

分布式系统与数据中心 (11篇)

Basilisk: Using Provenance Invariants to Automate Proofs of Undecidable Protocols

Authors: Tony Nuda Zhang and Keshav Singh, University of Michigan; Tej Chajed, UW-Madison; Manos Kapritsos, University of Michigan; Bryan Parno, Carnegie Mellon University

Deriving Semantic Checkers from Tests to Detect Silent Failures in Production Distributed Systems

Authors: Chang Lou and Dimas Shidqi Parikesit, University of Virginia; Yujin Huang, The Pennsylvania State University; Zhewen Yang and Senapati Diwangkara, Johns Hopkins University; Yuzhuo Jing, University of Michigan; Achmad Imam Kistijantoro, Bandung Institute of Technology; Ding Yuan, University of Toronto; Suman Nath, Microsoft Research; Peng Huang, University of Michigan

Picsou: Enabling Replicated State Machines to Communicate Efficiently

Authors: Reginald Frank and Micah Murray, UC Berkeley; Suyash Gupta, University of Oregon; Qibao Xu, Chawinphat Tankuranand, Junseo Yoo, and Natacha Crooks, UC Berkeley; Manos Kapritsos, University of Michigan

FineMem: Breaking the Allocation Overhead vs. Memory Waste Dilemma in Fine-Grained Disaggregated Memory Management

Authors: Xiaoyang Wang and Yongkun Li, University of Science and Technology of China; Kan Wu, Google; Wenzhe Zhu, Yuqi Li, and Yinlong Xu, University of Science and Technology of China

To PRI or Not To PRI, That's the question

Authors: Yun Wang, Shanghai Jiao Tong University; Liang Chen, Jie Ji, Xianting Tian, and Ben Luo, Alibaba Group; ZhiXiang Wei, Zhibai Huang, and Kialiang Xu, Shanghai Jiao Tong University; Kaihuan Peng, Kaijie Guo, Ning Luo, Guangjian Wang, Shengdong Dai, and Yibin Shen, Alibaba Cloud; Jiesheng Wu, Alibaba; Zhengwei Qi, Shanghai Jiao Tong University

Enabling Efficient GPU Communication over Multiple NICs with FuseLink

Authors: Zhenghang Ren, Yuxuan Li, Zilong Wang, Xinyang HUANG, Wenxue Li, Kaiqiang Xu, Xudong Liao, Yijun Sun, and Bowen Liu, Hong Kong University of Science and Technology; Han Tian, University of Science and Technology of China; Junxue Zhang, Hong Kong University of Science and Technology; Mingfei Wang, MetaX Integrated Circuits (Shanghai) Co., Ltd; Zhizhen Zhong, Massachusetts Institute of Technology; Guyue Liu, Peking University; Ying Zhang, Meta; Kai Chen, Hong Kong University of Science and Technology

Low End-to-End Latency atop a Speculative Shared Log with Fix-Ante Ordering

Authors: Shreesha Gopalakrishna Bhat, Tony Hong, Xuhao Luo, Jiyu Hu, Aishwarya Ganesan, and Ramnatthan Alagappan, University of Illinois Urbana-Champaign

Understanding Stragglers in Large Model Training Using What-if Analysis

Authors: Jinkun Lin, New York University; Ziheng Jiang, Zuquan Song, Sida Zhao, and Menghan Yu, ByteDance Seed; Zhanghan Wang, New York University; Chenyuan Wang, ByteDance Seed; Zuocheng Shi, Zhejiang University; Xiang Shi, ByteDance; Wei Jia, Zherui Liu, Shuguang Wang, Haibin Lin, and Xin Liu, ByteDance Seed; Aurojit Panda and Jinyang Li, New York University

Fork in the Road: Reflections and Optimizations for Cold Start Latency in Production Serverless Systems

Authors: Xiaohu Chai, Tsinghua University and Ant Group; Tianyu Zhou, Ant Group; Keyang Hu, Tsinghua University; Jianfeng Tan, Tiwei Bie, Anqi Shen, Dawei Shen, Qi Xing, Shun Song, Tongkai Yang, Le Gao, Feng Yu, and Zhengyu He, Ant Group; Dong Du and Yubin Xia, Shanghai Jiao Tong University; Kang Chen, Tsinghua University; Yu Chen, Quan Cheng Laboratory, Jinan, China, and Tsinghua University

Kamino: Efficient VM Allocation at Scale with Latency-Driven Cache-Aware Scheduling

Authors: David Domingo, Rutgers University; Hugo Barbalho and Marco Molinaro, Microsoft Research; Kuan Liu and Abhisek Pan, Microsoft; David Dion, Microsoft Azure; Thomas Moscibroda, Microsoft; Sudarsun Kannan, Rutgers University; Ishai Menache, Microsoft Research

Empowering Distributed Training with Sparsity-driven Data Synchronization

Authors: Zhuang Wang and Yuke Wang, Rice University; Zhaozhuo Xu, Stevens Institute of Technology; ; Jingyi Xi, Zhejiang University; Anshumali Shrivastava and T. S. Eugene Ng, Rice University

文件与存储系统 (5篇)

Stripeless Data Placement for Erasure-Coded In-Memory Storage

Authors: Jian Gao, Jiwu Shu, Bin Yan, and Yuhao Zhang, Tsinghua University; Keji Huang, Huawei Technologies Co., Ltd

PoWER Never Corrupts: Tool-Agnostic Verification of Crash Consistency and Corruption Detection

Authors: Hayley LeBlanc, University of Texas at Austin; Jacob R. Lorch and Chris Hawblitzel, Microsoft Research; Cheng Huang and Yiheng Tao, Microsoft; Nickolai Zeldovich, MIT CSAIL and Microsoft Research; Vijay Chidambaram, University of Texas at Austin

Fast and Synchronous Crash Consistency with Metadata Write-Once File System

Authors: Yanqi Pan, Wen Xia, Yifeng Zhang, Xiangyu Zou, and Hao Huang, Harbin institute of Technology, Shenzhen; Zhenhua Li, Tsinghua University; Chentao Wu, Shanghai Jiao Tong University

Decentralized, Epoch-based F2FS Journaling With Fine-grained Crash Recovery

Authors: Yaotian Cui and Zhiqi Wang, The Chinese University of Hong kong, China; Renhai Chen, College of Intelligence and Computing, Tianjin University, China; Zili Shao, The Chinese University of Hong Kong, China

Nyala: Decoupling data striping and redundancy grouping in cluster file systems

Authors: Sanjith Athlur and Timothy Kim, Carnegie Mellon University; Saurabh Kadekodi, Google; Francisco Maturana and Xavier Ramos, Carnegie Mellon University; Arif Merchant, Google; Rashmi Vinayak and Greg Ganger, Carnegie Mellon University

内核与操作系统 (10篇)

Extending Applications Safely and Efficiently

Authors: Yusheng Zheng, UC Santa Cruz; Tong Yu, eunomia-bpf Community; Yiwei Yang, UC Santa Cruz; Yanpeng Hu, ShanghaiTech University; Xiaozheng Lai, South China University of Technology; Dan Williams, Virginia Tech; Andrew Quinn, UC Santa Cruz

A Unified Hardware Performance Profiling Infrastructure to Measure and Manage Uncertainty

Authors: Ao Li, Marion Sudvarg, Zihan Li, Sanjoy Baruah, Chris Gill, and Ning Zhang, Washington University in St. Louis

Building Bridges: Safe Interactions with Foreign Languages through Omniglot

Authors: Leon Schuermann and Jack Toubes, Princeton University; Tyler Potyondy and Pat Pannuto, University of California San Diego; Mae Milano and Amit Levy, Princeton University

KRR: Efficient and Scalable Kernel Record Replay

Authors: Tianren Zhang, unaffiliated; Sishuai Gong and Pedro Fonseca, Purdue University

Debox: Enforcing Determinism on Untrusted Machine Code

Authors: Zachary Yedidia, Geoffrey Ramseyer, and David Mazieres, Stanford University

Disentangling the Dual Role of NIC Receive Rings

Authors: Boris Pismenny, EPFL and NVIDIA; Adam Morrison, Tel Aviv University; Dan Tsafrir, Technion

Preemptive Scheduling for Diverse XPUs using Multi-level Hardware Model

Authors: Weihang Shen, Mingcong Han, Jialong Liu, Rong Chen, and Haibo Chen, Shanghai Jiao Tong University

OS Rendering Service Made Parallel with Out-of-Order Execution and In-Order Commit

Authors: Yuanpei Wu and Dong Du, Shanghai Jiao Tong University; Chao Xu, Huawei Central Software Institute, Fields Lab; Yubin Xia, Shanghai Jiao Tong University; Ming Fu, Huawei Central Software Institute, Fields Lab; Yang Yu, Binyu Zang, and Haibo Chen, Shanghai Jiao Tong University

EMT: An OS Framework for New Memory Translation Architectures

Authors: Siyuan Chai, Jiyuan Zhang, Jongyul Kim, Alan Wang, Fan Chung, and Jovan Stojkovic, University of Illinois Urbana-Champaign; Weiwei Jia, University of Rhode Island; Dimitrios Skarlatos, Carnegie Mellon University; Josep Torrellas and Tianyin Xu, University of Illinois Urbana-Champaign

Tiered Memory Management Beyond Hotness

Authors: Jinshu Liu, Hamid Hadian, Hanchen Xu, and Huaicheng Li, Virginia Tech

调度与资源管理 (5篇)

Söze: One Network Telemetry Is All You Need For Per-flow Weighted Bandwidth Allocation at Scale

Authors: Weitao Wang, Rice University / Google; T. S. Eugene Ng, Rice University

Decouple and Decompose: Scaling Resource Allocation with DeDe

Authors: Zhiying Xu and Minlan Yu, Harvard University; Francis Y. Yan, University of Illinois Urbana-Champaign

Quantum Virtual Machines

Authors: Runzhou Tao, University of Maryland; Hongzheng Zhu and Jason Nieh, Columbia University

QOS: Quantum Operating System

Authors: Emmanouil Giortamis, Francisco Romão, Nathaniel Tornow, and Pramod Bhatotia, TU Munich

Scalio: Scaling up DPU-based JBOF Key-value Store with NVMe-oF Target Offload

Authors: Xun Sun, Mingxing Zhang, Yingdi Shan, Kang Chen, Jinlei Jiang, and Yongwei Wu, Tsinghua University

数据库系统 (5篇)

Tigon: A Distributed Database for a CXL Pod

Authors: Yibo Huang, Haowei Chen, Newton Ni, Vijay Chidambaram, Dixin Tang and Emmett Witchel, The University of Texas at Austin

Warbler: Speculative Distributed Transactions with Geo-Replication

Authors: Weihai Shen, Stony Brook University; Yang Cui, Google; Siddhartha Sen, Microsoft Research; Sebastian Angel, University of Pennsylvania; Shuai Mu, Stony Brook University

Quake: Adaptive Indexing for Vector Search

Authors: Jason Mohoney, Devesh Sarda, and Mengze Tang, University of Wisconsin, Madison; Shihabur Rahman Chowdhury and Anil Pacaci, Apple; Ihab F. Ilyas, University of Waterloo; Theodoros Rekatsinas, Apple; Shivaram Venkataraman, University of Wisconsin, Madison

Achieving Low-Latency Graph-Based Vector Search via Aligning Best-First Search Algorithm with SSD

Authors: Hao Guo and Youyou Lu, Tsinghua University

Skybridge: Bounded Staleness for Distributed Caches

Authors: Robert Lyerly and John Hugg, Meta Platforms, Inc.

隐私与安全 (4篇)

Astrolabe: Encrypted Semantic Search with High Accuracy

Authors: Jinhao Zhu, UC Berkeley; Liana Patel, Stanford University; Matei Zaharia and Raluca Ada Popa, UC Berkeley

Weave: Efficient and Expressive Oblivious Analytics at Scale

Authors: Mahdi Soleimani, Grace Jia, and Anurag Khandelwal, Yale University

Paralegal: Practical Static Analysis for Privacy Bugs

Authors: Justus Adam, Carolyn Zech, Livia Zhu, Sreshtaa Rajesh, Nathan Harbison, Mithi Jethwa, Will Crichton, Malte Schwarzkopf, and Shriram Krishnamurthi, Brown University

Carpet: Costs and Benefits of Implementing Containers on Microkernels

Authors: Till Miemietz, Matthias Hille, and Viktor Reusch, Barkhausen Institut; Lars Wrenger, Leibniz Universität Hannover; Jana Eisoldt, Barkhausen Institut; Jan Klötzke, Kernkonzept GmbH; Max Kurze and Adam Lackorzynski, TU Dresden; Michael Roitzsch and Hermann Härtig, Barkhausen Institut

其他 (1篇)

Title TBA

Authors:

Abstract: Emery Berger is a Professor of Computer Science at the University of Massachusetts Amherst, the flagship campus of the UMass system, and an Amazon Scholar at Amazon Web Services. At UMass, Professor Berger leads the PLASMA lab, whose research has led to numerous impactful software systems (see https://github.com/plasma-umass). Professor Berger is also the developer and sole maintainer of the influential CSrankings.org site, which has served over 3 million users. He served six years as an elected member of the SIGPLAN Executive Committee and a decade as Associate Editor of TOPLAS; he served as Program Chair for PLDI 2016 and co-Program Chair of ASPLOS 2021, and received the ACM SIGPLAN Distinguished Service Award in 2024. His honors include an NSF CAREER Award, Most Influential Paper Awards at OOPSLA, PLDI, and ASPLOS, five CACM Research Highlights, and Best Paper Awards at FAST, OOPSLA, SOSP, and OSDI; he is an ACM Fellow.