9:00 am–10:00 am
USENIX ATC '23 and OSDI '23 Joint Keynote Address
Title TBA
Ion Stoica, University of California, Berkeley
10:00 am–10:30 am
Break with Refreshments
10:30 am–10:45 am
Opening Remarks and Awards
Julia Lawall, Inria, and Dan Williams, Virginia Tech
10:45 am–10:55 am
Short Break
10:55 am–12:10 pm
Security and Privacy
Analysis and Optimization of Network I/O Tax in Confidential Virtual Machines
Dingji Li, Zeyu Mi, Chenhui Ji, Yifan Tan, Binyu Zang, Haibing Guan, and Haibo Chen, Shanghai Jiao Tong University
SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning
Junming Ma, Yancheng Zheng, Jun Feng, Derun Zhao, Haoqi Wu, Wenjing Fang, Jin Tan, Chaofan Yu, Benyu Zhang, and Lei Wang, Ant Group
Portunus: Re-imagining Access Control in Distributed Systems
Watson Ladd, Akamai; Marloes Venema, Radboud University/University of Wuppertal; Tanya Verma, Cloudflare, Inc
Searching Graphs
GLogS: Interactive Graph Pattern Matching Query At Large Scale
Longbin Lai, Alibaba Group; Yufan Yang, The Chinese University of Hong Kong, Shenzhen; Zhibin Wang, Nanjing University; Yuxuan Liu and Haotian Ma, The Chinese University of Hong Kong, Shenzhen; Sijie Shen, Bingqing Lyu, Xiaoli Zhou, Wenyuan Yu, and Zhengping Qian, Alibaba Group; Chen Tian and Sheng Zhong, Nanjing University; Yeh-Ching Chung, The Chinese University of Hong Kong, Shenzhen; Jingren Zhou, Alibaba Group
Cyclosa: Redundancy-Free Graph Pattern Mining via Set Dataflow
Chuangyi Gui, Xiaofei Liao, Long Zheng, and Hai Jin, Huazhong University of Science and Technology
SOWalker: An I/O-Optimized Out-of-Core Graph Processing System for Second-Order Random Walks
Yutong Wu, Zhan Shi, Shicai Huang, Zhipeng Tian, Pengwei Zuo, Peng Fang, Fang Wang, and Dan Feng, Huazhong University of Science and Technology
12:10 pm–1:40 pm
Conference Luncheon
1:40 pm–2:55 pm
Deduplication
Light-Dedup: A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems
Jiansheng Qiu, Harbin Institute of Technology, Shenzhen (Now work at Tsinghua University); Yanqi Pan, Harbin institute of Technology, Shenzhen; Wen Xia, Harbin Institute of Technology, Shenzhen; Xiaojia Huang, Harbin Institute of Technology,Shenzhen; Wenjun Wu and Xiangyu Zou, School of Computer Science and Technology Harbin Institute of Technology Shenzhen,China; Shiyi Li, Harbin institute of Technology, Shenzhen; Yu Hua, Huazhong University of Science and Technology
TiDedup: A New Distributed Deduplication Architecture for Ceph
Myoungwon Oh and Sungmin Lee, Samsung Electronics; Samuel Just, IBM; Young Jin Yu and Duckho Bae, Samsung Electronics; Sage Weil, Ceph Foundation; Sangyeun Cho, Samsung Electronics; Heon Y. Yeom, Seoul National University
LoopDelta: Embedding Locality-aware Opportunistic Delta Compression in Inline Deduplication for Highly Efficient Data Reduction
Yucheng Zhang, Nanchang University; Hong Jiang, UT Arlington; Dan Feng, Huazhong University of Science and Technology; Nan Jiang, East China Jiaotong University, Nanchang, China; Taorong Qiu and Wei Huang, Nanchang University
Structuring Graphs
TC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs
Yuke Wang and Boyuan Feng, University of California Santa Barbara; Zheng Wang and Guyue Huang, University of California, Santa Barbara; Yufei Ding, University of California Santa Barbara
Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale GNN Training
Jie Sun, Zhejiang University; Li Su, Alibaba Group; Zuocheng Shi, Zhejiang University; Wenting Shen, Alibaba Group; Zeke Wang, Zhejiang University; Lei Wang, Alibaba Group; Jie Zhang, Zhejiang University; Wenyuan Yu, Yong Li, and Jingren Zhou, Alibaba Group; Fei Wu, Zhejiang University
Bridging the Gap between Relational OLTP and Graph-based OLAP
Sijie Shen, Shanghai Jiao Tong University; Alibaba Group; Zihang Yao and Lin Shi, Shanghai Jiao Tong University; Lei Wang, Longbin Lai, Qian Tao, and Li Su, Alibaba Group; Rong Chen, Shanghai Jiao Tong University; Shanghai AI Laboratory; Wenyuan Yu, Alibaba Group; Haibo Chen and Binyu Zang, Shanghai Jiao Tong University; Jingren Zhou, Alibaba Group
2:55 pm–3:35 pm
Break with Refreshments
3:35 pm–5:05 pm
Placement and Fault Tolerance
Comosum: An Extensible, Reconfigurable, and Fault-Tolerant IoT Platform for Digital Agriculture
Gloire Rubambiza, Shiang-Wan Chin, Sachille Atapattu, Mueed Rehman, and José F. Martínez, Cornell University; Hakim Weatherspoon, Cornell University and Exotanium, Inc.
Oakestra: A Lightweight Hierarchical Orchestration Framework for Edge Computing
Giovanni Bartolomeo, Mehdi Yosofie, Simon Bäurle, Oliver Haluszczynski, Nitinder Mohan, and Jorg Ott, Technische Universität München
Explore Data Placement Algorithm for Balanced Recovery Load Distribution
Yingdi Shan, Zhongguancun Laboratory
Updating Code
Luci: Loader-based Dynamic Software Updates for Off-the-shelf Shared Objects
Bernhard Heinloth, Peter Wägemann, and Wolfgang Schröder-Preikschat, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)
MELF: Multivariant Executables for a Heterogeneous World
Dominik Töllner, Leibniz Universität Hannover; Christian Dietrich, Technische Universität Hamburg; Illia Ostapyshyn, Florian Rommel, and Daniel Lohmann, Leibniz Universität Hannover
APRON: Authenticated and Progressive System Image Renovation
Sangho Lee, Microsoft Research
zpoline: a system call hook mechanism based on binary rewriting
Kenichi Yasukata, Hajime Tazaki, and Pierre-Louis Aublin, IIJ Research Laboratory; Kenta Ishiguro, Hosei University
5:30 pm–7:00 pm
OSDI '23 Poster Session and Reception
Sponsored by Amazon
Would you like to share a provocative opinion, interesting preliminary work, or a cool idea that will spark discussion at this year's OSDI? The poster session is the perfect venue to introduce such new or ongoing work. Poster presenters will have the opportunity to discuss their work, get exposure, and receive feedback from other attendees during the in-person evening reception. View the Call for Posters and submit a proposal by June 7.
9:00 am–10:15 am
Serverless
Sponge: Fast Reactive Scaling for Stream Processing with Serverless Frameworks
Won Wook Song, Seoul National University; Taegeon Um, Samsung Research; Sameh Elnikety, Microsoft Research; Myeongjae Jeon, UNIST; Byung-Gon Chun, Seoul National University and FriendliAI
Deployed System: On-demand Container Loading in AWS Lambda
Marc Brooker, Chris Greenwood, and Mike Danilov, Amazon Web Services
Decentralized and Stateful Serverless Computing on the Internet Computer Blockchain
Maksym Arutyunyan, Andriy Berestovskyy, Adam Bratschi-Kaye, Ulan Degenbaev, Manu Drijvers, Islam El-Ashi, Stefan Kaestle, Roman Kashitsyn, Maciej Kot, Yvonne-Anne Pignolet, Rostislav Rumenov, Dimitris Sarlis, Alin Sinpalean, Alexandru Uta, Bogdan Warinschi, and Alexandra Zapuc, DFINITY
Troubleshooting and Measurement
Pinolo: Detecting Logical Bugs in Database Management Systems with Approximate Query Synthesis
Zongyin Hao and Quanfeng Huang, School of Informatics, Xiamen University; Chengpeng Wang, The Hong Kong University of Science and Technology; Jianfeng Wang, University of Southern California; Yushan Zhang, Tencent Inc.; Rongxin Wu, School of Informatics, Xiamen University; Charles Zhang, The Hong Kong University of Science and Technology
AutoARTS: Insights and Tools for Rootcausing Incidents in Microsoft Azure
Pradeep Dogga, UCLA; Chetan Bansal, Richard Costleigh, and Gopinath Jayagopal, Microsoft; Suman Nath, Microsoft Research; Xuchao Zhang, Microsoft
Avoiding the Ordering Trap in Systems Performance Measurement
Dmitry Duplyakin and Nikhil Ramesh, University of Utah; Carina Imburgia, University of Washington; Hamza Fathallah Al Sheikh, Semil Jain, Prikshit Tekta, Aleksander Maricq, Gary Wong, and Robert Ricci, University of Utah
10:15 am–10:45 am
Break with Refreshments
10:45 am–12:00 pm
Cloud and Microservices
AWARE: Automate Workload Autoscaling with Reinforcement Learning in Production Cloud Systems
Haoran Qiu and Weichao Mao, University of Illinois at Urbana-Champaign; Chen Wang and Hubertus Franke, IBM Thomas J. Watson Research Center; Alaa Youssef, IBM Thomas J. Watson Resource Center; Zbigniew T. Kalbarczyk, Tamer Başar, and Ravishankar K. Iyer, University of Illinois at Urbana-Champaign
Nodens: Enabling Resource Efficient and Fast QoS Recovery of Dynamic Microservice Applications in Datacenters
Jiuchen Shi, Quan Chen, Zhixin Tong, Hang Zhang, Kaihua Fu, and Minyi Guo, Shanghai Jiao Tong University
Lifting the veil on Meta’s microservice architecture: Analyses of topology and request workflows
Darby Huye, Meta & Tufts University; Yuri Shkuro, Meta; Raja Sambasivan, Tufts University
Distributed Storage
TectonicShift: A Composite Storage Fabric for Large-Scale ML Training
Mark Zhao, Stanford University; Satadru Pan, Meta; Niket Agarwal, NVIDIA; Zhaoduo Wen, David Xu, Anand Natarajan, Pavan Kumar, Shiva Shankar P, Ritesh Tijoriwala, Karan Asher, Hao Wu, Aarti Basant, Daniel Ford, Delia David, Nezih Yigitbasi, and Pratap Singh, Meta; Carole-Jean Wu, Facebook/ASU; Christos Kozyrakis, Stanford University
Calcspar: A Contract-Aware LSM-Tree for Cloud Storage with Low Latency Spikes
Yuanhui Zhou and Jian Zhou, WNLO, Huazhong University of Science and Technology; Shuning Chen, PingCAP, China; Peng Xu, Zhejiang Lab; Peng Wu, WNLO, Huazhong University of Science and Technology; Yanguang Wang, PingCAP，China; Xian Liu, PingCAP; Ling Zhan, Division of Information Science and Technology, Wenhua University, Wuhan, China; Jiguang Wan, Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology
Adaptive Online Cache Capacity Optimization via Lightweight Working Set Size Estimation at Scale
Rong Gu, Simian Li, Haipeng Dai, Hancheng Wang, and Yili Luo, Nanjing University; Bin Fan, Alluxio Inc.; Ran Ben Basat, University College London; Ke Wang, Meta Inc.; Zhenyu Song, Princeton University; Shouwei Chen and Beinan Wang, Alluxio Inc.; Yihua Huang and Guihai Chen, Nanjing University
12:00 pm–1:40 pm
Conference Luncheon
1:40 pm–2:55 pm
Hardware and Software for Security and Performance
SAGE: Software-based Attestation for GPU Execution
Andrei Ivanov and Benjamin Rothenberger, ETH Zurich; Arnaud Dethise and Marco Canini, KAUST; Torsten Hoefler and Adrian Perrig, ETH Zurich
Confidential Computing within an AI Accelerator
Kapil Vaswani, Stavros Volos, Cedric Fournet, Antonio Nino Diaz, and Ken Gordon, Microsoft Research; Balaji Vembu, Meta; Sam Webster, Microsoft Research; David Chisnall, Microsoft Research, University of Cambridge; Saurabh Kulkarni, Lucata Systems; Graham Cunningham and Richard Osborne, Graphcore; Dan Wilkinson, Imagination Technologies
Arbitor: A Numerically Accurate Hardware Emulation Tool for DNN Accelerators
Chenhao Jiang and Anand Jayarajan, University of Toronto / Vector Institute; Hao Lu, University of Toronto; Gennady Pekhimenko, University of Toronto / Vector Institute
Networking
oBBR: Optimize Retransmissions of BBR flows on the Internet
Pengqiang Bi, Mengbai Xiao, Dongxiao Yu, and Guanghui Zhang, Shandong University; Jian Tong, Jingchao Liu, and Yijun Li, BaishanCloud
Bridging the Gap between QoE and QoS in Congestion Control: A Large-scale Mobile Web Service Perspective
Jia Zhang, Yixuan Zhang, Enhuan Dong, Yan Zhang, Shaorui Ren, Zili Meng, and Mingwei Xu, Tsinghua University; Xiaotian Li, Zongzhi Hou, and Zhicheng Yang, Meituan; Xiaoming Fu, Institut für Informatik, Georg-August-Universität Göttingen, Göttingen, Germany
FarReach: Write-back Caching in Programmable Switches
Siyuan Sheng and Huancheng Puyang, The Chinese University of Hong Kong; Qun Huang, Peking University; Lu Tang, Xiamen University; Patrick P. C. Lee, The Chinese University of Hong Kong
2:55 pm–3:20 pm
Break with Refreshments
3:20 pm–4:35 pm
Memory-Related Hardware and Software
CXL-ANNS: Software-Hardware Collaborative Memory Disaggregation and Computation for Billion-Scale Approximate Nearest Neighbor Search
Junhyeok Jang, Hanjin Choi, Hanyeoreum Bae, and Seungjun Lee, KAIST; Miryeong Kwon and Myoungsoo Jung, Panmnesia Research and KAIST
Overcoming the Memory Wall with CXL-Enabled SSDs
Shao-Peng Yang, Syracuse University; Minjae Kim, DGIST; Sanghyun Nam, Soongsil University; Juhyung Park, DGIST; Jin-yong Choi and Eyee Hyun Nam, FADU; Eunji Lee, Soongsil University; Sungjin Lee, DGIST; Bryan S. Kim, Syracuse University
STYX: Exploiting SmartNIC Capability to Reduce Datacenter Memory Tax
Houxiang Ji and Yan Sun, University of Illinois Urbana-Champaign; Mark Mansi, University of Wisconsin-Madison; Yifan Yuan, Intel Labs; Jinghan Huang and Reese Kuper, University of Illinois Urbana-Champaign; Michael M. Swift, University of Wisconsin-Madison; Nam Sung Kim, University of Illinois Urbana-Champaign
Deployed Networking
Change Management in Physical Network Lifecycle Automation
Mohammad Al-Fares, Virginia Beauregard, Kevin Grant, Angus Griffith, Quan Leng, and Alexander Lin, Google; Zhuotao Liu, Tsinghua University; Bill Martinusen, Formerly Google; Nikil Mehta, Jeffrey Mogul, Andrew Narver, and Anshul Nigham, Google; Sean Smith, Databricks; Amin Vahdat, Google
AAsclepius: Monitoring, Diagnosing, and Detouring at the Internet Peering Edge
Kaicheng Yang, Yuanpeng Li, Sheng Long, Tong Yang, Ruijie Miao, and Yikai Zhao, Peking University; Chaoyang Ji, Penghui Mi, Guodong Yang, Qiong Xie, Hao Wang, Yinhua Wang, Bo Deng, Zhiqiang Liao, Chengqiang Huang, Yongqiang Yang, Xiang Huang, Wei Sun, and Xiaoping Zhu, Huawei Cloud Computing Technologies Co., Ltd.
Deploying User-space TCP at Cloud Scale: Three Easy Pieces
Lingjun Zhu, Yifan Shen, Erci Xu, Bo Shi, Ting Fu, Shu Ma, Shuguang Chen, Zhongyu Wang, Haonan Wu, Zhendan Yang, Zhongqing Chen, and Rong Liu, Alibaba Inc.; Chao Shi and Jiaji Zhu, Alibaba Group; Jiesheng Wu, Xingyu Liao, Wei Lin, and Yijun hou, Alibaba Inc.
4:35 pm–4:45 pm
Short Break
4:45 pm–5:35 pm
Key-Value Stores
RubbleDB: CPU-Efficient Replication with NVMe-oF
Haoyu Li, Sheng Jiang, and Chen Chen, Columbia University; Ashwini Raina, Princeton University; Xingyu Zhu, Changxu Luo, and Asaf Cidon, Columbia University
Distributed Transactions at Scale in DynamoDB
Akshat Vig, Amazon Web Services; Somasundaram Perianayagam, Amazon.com; Doug Terry, Prithvi Ramanathan, and Alex Keyes, Amazon Web Services
Security: Attacks
Prefix Siphoning: Exploiting LSM-Tree Range Filters For Information Disclosure
Adi Kaufman, Moshik Hershcovitch, and Adam Morrison, Tel Aviv University
EPF: Evil Packet Filter
Di Jin, Vaggelis Atlidakis, and Vasileios P. Kemerlis, Brown University
6:00 pm–7:30 pm
USENIX ATC '23 Poster Session and Reception
The USENIX ATC '23 poster session and reception will feature posters by authors presenting their work in person at the conference. USENIX ATC '23 authors can find out how to present their work on the Instructions for Presenters webpage.
9:00 am–10:15 am
Virtual Machines
Translation Pass-Through for Near-Native Paging Performance in VMs
Shai Bergman and Mark Silberstein, Technion; Takahiro Shinagawa, The University of Tokyo; Peter Pietzuch and Lluís Vilanova, Imperial College London
Efficient Memory Overcommitment for I/O Passthrough Enabled VMs via Fine-grained Page Meta-data Management
Yaohui Wang, Ben Luo, and Yibin Shen, Alibaba Group
LPNS: Scalable and Latency-Predictable Local Storage Virtualization for Unpredictable NVMe SSDs in Clouds
Bo Peng, Guo Cheng, Jianguo Yao, and Haibing Guan, Shanghai Jiao Tong University
Persistent Memory
P2CACHE: Exploring Tiered Memory for In-Kernel File Systems Caching
Zhen Lin, State University of New York at Binghamton; Lingfeng Xiang and Jia Rao, The University of Texas at Arlington; Hui Lu, State University of New York at Binghamton
Revisiting Secondary Indexing in LSM-based Storage Systems with Persistent Memory
Jing Wang, Youyou Lu, Qing Wang, Yuhao Zhang, and Jiwu Shu, Tsinghua University
Zhuque: Failure is Not an Option, it’s an Exception
George Hodgkins, University of Colorado Boulder; Yi Xu, University of California, San Diego; Steve Swanson, University of California - San Diego; Joseph Izraelevitz, University of Colorado, Boulder
10:15 am–10:45 am
Break with Refreshments
10:45 am–12:00 pm
Offloading and Scheduling
EnvPipe: Performance-preserving DNN Training Framework for Saving Energy
Sangjin Choi and Inhoe Koo, KAIST; Jeongseob Ahn, Ajou University; Myeongjae Jeon, UNIST; Youngjin Kwon, KAIST
Decentralized Application-Level Adaptive Scheduling for Multi-Instance DNNs on Open Mobile Devices
Hsin-Hsuan Sung, North Carolina State University; Jiexiong Guan and Wei Niu, William & Mary; Jou-An Chen, North Carolina State University; Bin Ren, William & Mary; Xipeng Shen, North Carolina State University
UnFaaSener: Latency and Cost Aware Offloading of Functions from Serverless Platforms
Ghazal Sadeghian and Mohamed Elsakhawy, University of British Columbia; Mohanna Shahrad, McGill University; Joe Hattori, The University of Tokyo; Mohammad Shahrad, University of British Columbia
Kernel and Concurrency
LLFree: Scalable and Optionally-Persistent Page-Frame Allocation
Lars Wrenger, Florian Rommel, and Alexander Halbuer, Leibniz Universität Hannover; Christian Dietrich, Technische Universität Hamburg; Daniel Lohmann, Leibniz Universität Hannover
SingularFS: A Billion-Scale Distributed File System Using a Single Metadata Server
Hao Guo, Youyou Lu, Wenhao Lv, Xiaojian Liao, Shaoxun Zeng, and Jiwu Shu, Tsinghua University
The Hitchhiker's Guide to Operating Systems
Yanyan Jiang, Nanjing University
12:00 pm–1:40 pm
Lunch (on your own)
1:40 pm–2:55 pm
Optimizing ML
Accelerating Distributed MoE Training and Inference with Lina
Jiamin Li, City University of Hong Kong; Yimin Jiang and Yibo Zhu, ByteDance Inc.; Cong Wang, City University of Hong Kong; Hong Xu, The Chinese University of Hong Kong
SmartMoE: Efficiently Training Sparsely-Activated Models through Combining Static and Dynamic Parallelization
Mingshu Zhai, Jiaao He, Zixuan Ma, Zan Zong, Runqing Zhang, and Jidong Zhai, Tsinghua University
AthenaRL: Distributed Reinforcement Learning with Dataflow Fragments
Huanzhou Zhu and Bo Zhao, Imperial College; Gang Chen, Weifeng Chen, Yijie Chen, and Shi Liang, Huawei; Yaodong Yang, Institute for AI, Peking University; Peter Pietzuch, Imperial College London; Lei Chen, HongKong University of Science and Technology
2:55 pm–3:35 pm
Break with Refreshments
3:35 pm–4:40 pm
GPU
Beware of Fragmentation: Scheduling GPU-Sharing Workloads with Fragmentation Gradient Descent
Qizhen Weng and Lingyun Yang, Hong Kong University of Science and Technology; Yinghao Yu, Alibaba Group; Wei Wang, Hong Kong University of Science and Technology; Xiaochuan Tang, Guodong Yang, and Liping Zhang, Alibaba Group
Towards Iterative Relational Algebra on the GPU
Ahmedur Rahman Shovon and Thomas Gilray, University of Alabama at Birmingham; Kristopher Micinski, Syracuse University; Sidharth Kumar, University of Alabama at Birmingham
VectorVisor: A Binary Translation Scheme for Throughput-Oriented GPU Acceleration
Samuel Ginzburg, Princeton University; Mohammad Shahrad, University of British Columbia; Michael J. Freedman, Princeton University
4:40 pm–4:50 pm
Closing Remarks
Julia Lawall, Inria, and Dan Williams, Virginia Tech