Abstract
A wide range of research areas – from natural language processing to computer vision and
software engineering – have been (or are being) revolutionized by machine learning and
artificial intelligence. Each of these areas went through an inflection point where they
transitioned from ML as one of many approaches to ML becoming a predominant approach of the
field. No example symbolizes this better than the AlexNet paper from 2012, which
fundamentally transformed the field of computer vision.
Computer systems remain a notable exception. In this talk, I will discuss emerging trends in
the ML for Systems domain, how systems differ from these other areas, and what an "AlexNet
Moment" for systems might look like. Along the way, I will describe a framework for
categorizing work in the field and discuss emerging research problems and opportunities.
Bio
Martin Maas is a research scientist at Google DeepMind, where he is working on new
approaches to leverage artificial intelligence for solving computer systems problems. His
research has been deployed in a range of Google systems and products, including Google
Compute Engine, TCMalloc, Pixel phones and the TPU compiler. His work has received multiple
recognitions, including an ASPLOS Best Paper Award, an IEEE Micro Top Pick, a SIGPLAN
Research Highlight, and a CACM Research Highlight. He has been active in several leadership
roles in the community, including as General Chair of ISMM 2025, WACI Chair and Vice Program
Chair at ASPLOS 2025, Program Chair of ISMM 2020, and as one of the co-organizers of the ML
for Systems workshop at NeurIPS. He also co-leads Google’s involvement in the free and open
RISC-V instruction set architecture. Martin holds a Ph.D. from the University of California
at Berkeley and a B.A. from the University of Cambridge, both in Computer Science.
Program
ASPLOS 2025 will be running concurrently with EuroSys. The ASPLOS program is available here.Monday 31st March
18:00-19:30 (Monday) Welcome Reception & EuroSys Poster Session - Location: Catering area
Tuesday 1st April
08:00-8:45 Registration and Welcome Coffee - Location: Catering area
8:45-9:00 Opening - Location: Rotterdam Hall 1
9:00-10:00 ASPLOS + EuroSys 2025 Joint Keynote 1 by Martin Maas (Google DeepMind) -- Location: Rotterdam Hall 1 - Chair: Chris Rossbach
10:00-10:30 Coffee Break - Location: Catering area
10:30-11:00 Award ceremony - Location: Rotterdam Hall 1
11:00-12:00 ASPLOS + EuroSys 2025 Joint Keynote 2 by Gernot Heiser (Univ. of New South Wales) - Location: Rotterdam Hall 1 - Chair: Haibo Chen
Abstract
Half a century after PSOS, the first attempts to prove an operating system (OS) secure, OS
faults remain a major threat to computer systems security. A major step forward was the
verification of the seL4 microkernel, the first proof of implementation correctness of anOS
kernel. Over the next 4 years this proof was extended to the binary code, proofs of security
enforcement, and sound and complete worst-case execution-time analysis. The proofs now cover
4 ISAs.
Yet, 15 years later, there is still no provably secure OS. While seL4 has been successfully
deployed in defence and civilian security- and safety-critical systems, it is a microkernel
that mostly guarantees process isolation without providing the application-oriented services
expected from an OS. This not only makes seL4 difficult to deploy, but means that there is
limited assurance that a system built on top is secure in any real sense.
Why has seL4 not been leveraged into a secure OS? In this talk I will explore some of the
reasons behind this disappointing state of affairs, and what can be done about it.
Specifically I will discuss our current work on LionsOS, a new seL4-based OS targeting the
embedded/cyberphysical domain, and designed to be verifiable. I will also discuss more
speculative, early-stage work towards a provably secure, general-purpose OS.
Bio
Gernot Heiser is Scientia (distinguished) Professor and John Lions Chair of Operating
Systems
at UNSW Sydney, where he leads the Trustworthy Systems research group. His research interest
are in operating systems, real-time systems, security and safety. His research vision is to
completely change the cybersecurity game, from playing catch-up with attackers, to making
computer systems provably secure and safe. With his team he pioneered the large-scale formal
verification of systems code, specifically the design, implementation and formal
verification of the seL4 microkernel; this work was recognised with an ACM SIGOPS Hall of
Fame Award and the ACM Software System Award.
Heiser's former company Open Kernel Labs, acquired by General Dynamics in 2012, marketed the
OKL4 microkernel, which shipped on billions of mobile wireless chips and has been deployed
on the secure enclave of iOS devices. He presently serves as Chief Scientist of Neutrality,
and Chairman of the seL4 Foundation. Gernot is a Fellow of the ACM, the IEEE, Engineers
Australia, the Australian Academy of Technology and Engineering (ATSE) and the Royal Society
of New South Wales (RSN) and a Member of the German Academy of Sciences Leopoldina. He is
also an ACM Distinguished Lecturer and an IEEE Distinguished Visitor.
Why has seL4 not been leveraged into a secure OS? In this talk I will explore some of the
reasons behind this disappointing state of affairs, and what can be done about it.
Specifically I will discuss our current work on LionsOS, a new seL4-based OS targeting the
embedded/cyberphysical domain, and designed to be verifiable. I will also discuss more
speculative, early-stage work towards a provably secure, general-purpose OS.
12:00-13:30 Lunch - Location: Catering area
13:30-15:10
| Session-1.1: Operating Systems - Location: Mees - Chair: Jinyu Gu | Title | Authors |
|---|---|---|
| Empowering WebAssembly with Thin Kernel Interfaces | Arjun Ramesh (Carnegie Mellon University), Tianshu Huang (Carnegie Mellon University), Ben Titzer (CMU), Anthony Rowe (Carnegie Mellon University) | |
| Revealing the Unstable Foundations of eBPF-Based Kernel Extensions | Shawn Zhong (University of Wisconsin-Madison), Jing Liu (Microsoft Research), Andrea Arpaci-Dusseau (UW-Madison), Remzi Arpaci-Dusseau (University of Wisconsin–Madison) | |
| eNetSTL: Towards an In-kernel Library for High-Performance eBPF-based Network Functions | Bin Yang (Southeast University), Dian Shen (Southeast University), Junxue Zhang (Hong Kong University of Science and Technology), Hanlin Yang (Southeast University), Lunqi Zhao (Southeast University), Beilun Wang (Southeast University), Guyue Liu (Peking Univeristy), Kai Chen (Hong Kong University of Science and Technology) | |
| CRAVE: Analyzing Cross-Resource Interaction to Improve Energy Efficiency in Systems-on-Chip | Dipayan Mukherjee (University of Illinois, Urbana-Champaign), Sam Hachem (University of Illinois, Urbana-Champaign), Jeremy Bao (University of Illinois, Urbana-Champaign), Curtis Madsen (Sandia National Labs), Tian Ma (Sandia National Labs), Saugata Ghose (University of Illinois Urbana-Champaign), Gul Agha (University of Illinois at Urbana-Champaign) | |
| Efeu: generating efficient, verified, hybrid hardware/software drivers for I2C devices | Daniel Schwyn (ETH Zurich), Zikai Liu (ETH Zurich), Timothy Roscoe (ETH Zurich) |
| Session-1.2: LLM Serving - Location: Rotterdam hall 1B - Chair: Thaleia Doudali | Title | Authors |
|---|---|---|
| CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion | Jiayi Yao (Chinese University of Hong Kong (Shenzhen)), Hanchen Li (University of Chicago), Yuhan Liu (University of Chicago), Siddhant Ray (University of Chicago), Yihua Cheng (University of Chicago), Qizheng Zhang (Stanford University), Kuntai Du (University of Chicago), Shan Lu (Microsoft Research and University of Chicago), Junchen Jiang (University of Chicago) | |
| DeltaZip: Efficient Serving of Multiple Full-Model-Tuned LLMs | Xiaozhe Yao (ETH Zurich), Qinghao Hu (MIT), Ana Klimovic (ETH Zurich) | |
| Fast State Restoration in LLM Serving with HCache | Shiwei Gao (Tsinghua University), Youmin Chen (Tsinghua University), Jiwu Shu (Tsinghua University) | |
| Stateful Large Language Model Serving with Pensieve | Lingfan Yu (New York University), Jinkun Lin (New York University), Jinyang Li (New York University) | |
| SkyServe: Serving AI Models across Regions and Clouds with Spot Instances | Ziming Mao (UC Berkeley), Tian Xia (UC Berkeley), Zhanghao Wu (UC Berkeley), Wei-Lin Chiang (UC Berkeley), Tyler Griggs (UC Berkeley), Romil Bhardwaj (UC Berkeley), Zongheng Yang (UC Berkeley), Scott Shenker (ICSI AND UC Berkeley), Ion Stoica (UC Berkeley) |
15:10-15:40 Coffee Break
15:40-17:00
| Session-2.1: Fault Tolerance and Consensus - Location: Mees - Chair: Diogo Behrens | Title | Authors |
|---|---|---|
| RoboRebound: Multi-Robot System Defense with Bounded-Time Interaction | Neeraj Gandhi (University of Pennsylvania), Yifan Cai (University of Pennsylvania), Andreas Haeberlen (University of Pennsylvania), Linh Thi Xuan Phan (University of Pennsylvania) | |
| Achilles: Efficient TEE-Assisted BFT Consensus via Rollback Resilient Recovery | Jianyu Niu (Southern University of Science and Technology), Guanlong Wu (Southern University of Science and Technology), Shengqi Liu (Southern University of Science and Technology.), Xiaoqing Wen (University of British Columbia), Jiangshan Yu (The University of Sydney), Yinqian Zhang (Southern University of Science and Technology (SUSTech)) | |
| ParallelEVM: Operation-Level Concurrent Transaction Execution for EVM-Compatible Blockchains | Haoran Lin (Zhejiang University), Hang Feng (Zhejiang University), Yajin Zhou (Zhejiang University), Lei Wu (Zhejiang University) | |
| Ladon: High-Performance Multi-BFT Consensus via Dynamic Global Ordering | Hanzheng Lyu (University of British Columbia), Shaokang Xie (Southern University of Science and Technology), Jianyu Niu (Southern University of Science and Technology), Chen Feng (University of British Columbia), Yinqian Zhang (Southern University of Science and Technology), Ivan Beschastnikh (University of British Columbia) |
| Session-2.2: LLM Serving - Location: Rotterdam hall 1A - Chair: Luo Mai | Title | Authors |
|---|---|---|
| SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs | Ruibo FAN (Data Science and Analytics Thrust, HKUST(GZ)), Xiangrui YU (Data Science and Analytics Thrust, HKUST(GZ)), Peijie Dong (Data Science and Analytics Thrust, HKUST(GZ)), Zeyu Li (Data Science and Analytics Thrust, HKUST(GZ)), Gu Gong (Data Science and Analytics Thrust, HKUST(GZ)), QIANG WANG (Harbin Institute of Technology (Shenzhen)), Wei Wang (Hong Kong University of Science and Technology), Xiaowen Chu (Data Science and Analytics Thrust, HKUST(GZ)) | |
| Empower Vision Applications with LoRA LMM | Liang Mi (Nanjing University), Weijun Wang (Institute for AI Industry Research (AIR), Tsinghua University), Wenming Tu (Institute for AI Industry Research (AIR), Tsinghua University), Qingfeng He (Institute for AI Industry Research (AIR), Tsinghua University), Kui Kong (Institute for AI Industry Research (AIR), Tsinghua University), Xinyu Fang (Institute for AI Industry Research (AIR), Tsinghua University), Yazhu Dong (Institute for AI Industry Research (AIR), Tsinghua University), Yikang Zhang (Nanjing University), Yuanchun Li (Institute for AI Industry Research (AIR), Tsinghua University), Meng Li (Nanjing University), Haipeng Dai (Nanjing University), Guihai Chen (Nanjing University), Yunxin Liu (Institute for AI Industry Research (AIR), Tsinghua University), Weijun Wang (Tsinghua University) | |
| T-MAC: CPU Renaissance via Table Lookup for Low-Bit LLM Deployment on Edge | Jianyu Wei (USTC and Microsoft Research), Shijie Cao (Microsoft Research), Ting Cao (Microsoft Research), Lingxiao Ma (Microsoft Research), Lei Wang (UCAS and Microsoft Research), Yanyong Zhang (University of Science and Technology of China), Mao Yang (Microsoft Research) | |
| Samoyeds: Accelerating MoE Models with Structured Sparsity Leveraging Sparse Tensor Cores | Chenpeng Wu (Shanghai Jiao Tong University), Qiqi Gu (Shanghai Jiao Tong University), Heng Shi (Shanghai Enflame Technology Co.Ltd; Shanghai Jiao Tong University), Jianguo Yao (Shanghai Jiao Tong University), Haibing Guan (Shanghai Jiao Tong University) |
17:00-17:30 Coffee break
17:30-18:30 General assembly
Wednesday 2nd April
08:30-09:00 Registration and Welcome Coffee - Location: Catering area
09:00-10:40
| Session-3.1: Distributed Systems - Location: Rotterdam hall 1A - Chair: Yerom-David Bromberg | Title | Authors |
|---|---|---|
| Collaborative Text Editing with Eg-walker: Better, Faster, Smaller | Joseph Gentle (Independent), Martin Kleppmann (University of Cambridge) | |
| Themis: Finding Imbalance Failures in Distributed File Systems via a Load Variance Model | Yuanliang Chen (Tsinghua University), Fuchen Ma (Tsinghua University), Yuanhang Zhou (Tsinghua University), Zhen Yan (Tsinghua University), Qing Liao (Harbin Institute of Technology), Yu Jiang (Tsinghua University) | |
| Moko: Marrying Python with Big Data Systems | Ke Meng (Alibaba Group), Tao He (Alibaba Group), Sijie Shen (Alibaba Group), Lei Wang (Alibaba Group), Wenyuan Yu (Alibaba Group), Jingren Zhou (Alibaba Group) | |
| Pegasus: Transparent and Unified Kernel-Bypass Networking for Fast Local and Remote Communication | Dinglan Peng (Purdue University), Congyu Liu (Purdue University), Tapti Palit (Purdue University), Anjo Vahldiek-Oberwagner (Intel Labs), Mona Vij (Intel Labs), Pedro Fonseca (Purdue University) | |
| Multi-Grained Specifications for Distributed System Model Checking and Verification | Lingzhi Ouyang (Nanjing University), Xudong Sun (University of Illinois Urbana-Champaign), Ruize Tang (Nanjing University), Yu Huang (Nanjing University), Madhav Jivrajani (University of Illinois Urbana-Champaign), Xiaoxing Ma (Nanjing University), Tianyin Xu (University of Illinois Urbana-Champaign) |
| Session-3.2: Network/Network Congestion Control - Location: Diamond - Chair: Kai Chen | Title | Authors |
|---|---|---|
| Enabling Virtual Priority in Data Center Congestion Control | Zhaochen Zhang (Nanjing University), Feiyang Xue (Nanjing University), Keqiang He (Shanghai Jiao Tong University), Zhimeng Yin (City University of Hong Kong), Gianni Antichi (Politecnico Milano & Queen Mary University of London), Jiaqi Gao (Unaffiliated), Yizhi Wang (Nanjing University), Rui Ning (Nanjing University), Haixin Nan (Nanjing University), Xu Zhang (Nanjing University), Peirui Cao (Nanjing University), Xiaoliang Wang (Nanjing University), Wanchun Dou (Nanjing University), Guihai Chen (Nanjing University), Chen Tian (Nanjing University) | |
| Achieving Fairness Generalizability for Learning-based Congestion Control with Jury | Han Tian (University of Science and Technology of China), Xudong Liao (Hong Kong University of Science and Technology), Decang Sun (Hong Kong University of Science and Technology), Chaoliang Zeng (BitIntelligence), Yilun Jin (The Hong Kong University of Science and Technology), Junxue Zhang (Hong Kong University of Science and Technology), Xinchen Wan (Hong Kong University of Science and Technology), Zilong Wang (Hong Kong University of Science and Technology), Yong Wang (Hong Kong University of Science and Technology), Kai Chen (Hong Kong University of Science and Technology) | |
| Introspective Congestion Control for Consistent High Performance | Wanchun Jiang (School of Computer Science and Engineering, Central South University), Haoyang Li (School of Computer Science and Engineering, Central South University), Jia Wu (School of Computer Science and Engineering, Central South University), Kai Wang (School of Computer Science and Engineering, Central South University), Fengyuan Ren (Department of Computer Science and Technology, Tsinghua University), Jianxin Wang (School of Computer Science and Engineering, Central South University) | |
| Fork: A Dual Congestion Control Loop for Small and Large Flows in Datacenters | Yuan Liu (Tianjin University), Wenxin Li (Tianjin University & Huaxiahaorui Technology (Tianjin) Co., Ltd.), Yulong Li (Tianjin University), Lide Suo (Tianjin University), Xuan Gao (Tianjin University), Xin Xie (Tianjin University), Sheng Chen (Tianjin University & Huaxiahaorui Technology (Tianjin) Co., Ltd.), Ziqi Fan (Tianjin University), Wenyu Qu (Tianjin University), Guyue Liu (Peking University) | |
| Marlin: Enabling High-Throughput Congestion Control Testing in Large-Scale Networks | Yanqing Chen (Nanjing University), Li Wang (Nanjing University), Jingzhi Wang (Nanjing University), Songyue Liu (Nanjing University), Keqiang He (Shanghai Jiao Tong University), Jian Wang (Nanjing University), Xiaoliang Wang (Nanjing University), Wanchun Dou (Nanjing University), Guihai Chen (Nanjing University), Chen Tian (Nanjing University) |
10:40-11:10 Coffee break & Poster session (Wednesday afternoon presentations) - Location: Catering area
11:10-12:30
| Session-4.1: ML Systems - Location: Rotterdam Hall 1A - Chair: Yubin Xia | Title | Authors |
|---|---|---|
| LOFT: A Lock-free and Adaptive Learned Index with High Scalability for Dynamic Workloads | Yuxuan Mo (Huazhong University of Science and Technology), Yu Hua (Huazhong University of Science and Technology) | |
| MetaHG: Enhancing HGNN Systems Leveraging Advanced Metapath Graph Abstraction | Haiheng He (Huazhong University of Science and Technology), Haifeng Liu (Huazhong University of Science and Technology), Long Zheng (Huazhong University of Science and Technology), Yu Huang (Huazhong University of Science and Technology), Xinyang Shen (Huazhong University of Science and Technology), Wenkan Huang (Huazhong University of Science and Technology), Shuaihu Cao (Huazhong University of Science and Technology), XIAOFEI LIAO (Huazhong University of Science and Technology), Hai Jin (Huazhong University of Science and Technology), Jingling Xue (University of New South Wales) | |
| Flex: Fast, Accurate DNN Inference on Low-Cost Edges Using Heterogeneous Accelerator Execution | Tanmoy Sen (University of Virginia), Haiying Shen (University of Virginia), Anand Iyer (Georgia Tech) | |
| A House United Within Itself: SLO-Awareness for On-Premises Containerized ML Inference Clusters via Faro | Beomyeol Jeon (University of Illinois Urbana-Champaign), Chen Wang (IBM Research), Diana Arroyo (IBM Research), Alaa Youssef (IBM Research), Indranil Gupta (University of Illinois Urbana-Champaign) |
| Session-4.2: GPU Systems - Location: Mees - Chair: Yiming Zhang | Title | Authors |
|---|---|---|
| Comprehensive Deadlock Prevention for GPU Collective Communication | LiChen Pan (School of Computer Science, Peking University), Juncheng Liu (OneFlow Inc.), Yongquan Fu (Science and Technology Laboratory of Parallel and Distributed Processing; College of Computer, National University of Defense Technology, Changsha, Hunan province, China), Jinhui Yuan (OneFlow Inc.), Rongkai Zhang (None), PengZe Li (School of Computer Science, Peking University), Zhen Xiao (School of Computer Science, Peking University) | |
| Jupiter: Pushing the Speed and Scalability Limitations for Subgraph Matching on Multi-GPUs | Zhiheng Lin (Institute of Computing Technology, Chinese Academy of Sciences), Ke Meng (Institute of Computing Technology), Changjie Xu (Institute of Computing Technology, Chinese Academy of Sciences), Weichen Cao (Institute of Computing Technology, Chinese Academy of Sciences), Guangming Tan (Institute of Computing Technology) | |
| Improving GPU Sharing Performance through Adaptive Bubbleless Spatial-Temporal Sharing | Shulai Zhang (Shanghai Jiao Tong University), Quan Chen (Shanghai Jiao Tong University), Weihao Cui (Shanghai Jiao Tong University), Han Zhao (Shanghai Jiao Tong University), Chunyu Xue (Shanghai Jiao Tong University), Zhen Zheng (Microsoft), Wei Lin (Alibaba Group), Minyi Guo (Shanghai Jiao Tong University) | |
| Multiplexing Dynamic Deep Learning Workloads with SLO-awareness in GPU Clusters | Wenyan Chen (University of Macau; Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences), Chengzhi Lu (Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences; University of Chinese Academy of Sciences; University of Macau), Huanle Xu (University of Macau, Macau SAR, China), Kejiang Ye (Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences), ChengZhong Xu (University of Macau) |
12:30-14:00 Lunch - Location: Catering area
14:00-15:40
| Session-5.1: Graph and Stream processing - Location: Mees - Chair: Baptiste Lepers | Title | Authors |
|---|---|---|
| BINGO: Radix-based Bias Factorization for Random Walk on Dynamic Graphs | Pinhuan Wang (Rutgers, The State University of New Jersey), Chengying Huan (Nanjing University), Zhibin Wang (Nanjing University), Chen Tian (Nanjing University), Yuede Ji (The University of Texas at Arlington), Hang Liu (Rutgers, The State University of New Jersey) | |
| OHMiner: An Overlap-centric System for Efficient Hypergraph Pattern Mining | Hao Qi (Huazhong University of Science and Technology), Kang Luo (Huazhong University of Science and Technology), Ligang He (University of Warwick), Yu Zhang (Huazhong University of Science and Technology), Minzhi Cai (Huazhong University of Science and Technology), Jingxin Dai (Huazhong University of Science and Technology), Bingsheng He (National University of Singapore), Hai Jin (Huazhong University of Science and Technology), Zhan Zhang (Zhejiang Lab, China), Jin Zhao (Huazhong University of Science and Technology), Hengshan Yue (Jilin University), Hui Yu (Huazhong University of Science and Technology), Xiaofei Liao (Huazhong University of Science and Technology) | |
| Impeller: Stream Processing on Shared Logs | Zhiting Zhu (Lepton AI), Zhipeng Jia (Google), Newton Ni (University of Texas at Austin), Dixin Tang (UT Austin), Emmett Witchel (UT Austin) | |
| CAPSys: Contention-aware task placement for data stream processing | Yuanli Wang (Boston University), Lei Huang (Boston University), Zikun Wang (Boston University), Vasiliki Kalavri (Boston University), Ibrahim Matta (Boston University) | |
| NeuStream: Bridging Deep Learning Serving and Stream Processing | Haochen Yuan (Peking University), Yuanqing Wang (Peking University and Microsoft Research), Wenhao Xie (Peking University), Yu Cheng (Peking University and Microsoft Research), Ziming Miao (Microsoft Research), Lingxiao Ma (Microsoft Research), Jilong Xue (Microsoft Research), Zhi Yang (Peking University) |
| Session-5.2: Virtualization - Location: Penn - Chair: Reto Achermann | Title | Authors |
|---|---|---|
| Towards VM Rescheduling Optimization Through Deep Reinforcement Learning | Xianzhong Ding (University of California, Merced), Yunkai Zhang (University of California, Berkeley), Binbin Chen (ByteDance), Donghao Ying (UC Berkeley), Tieying Zhang (ByteDance), Jianjun Chen (Bytedance), Lei Zhang (ByteDance), Alberto Cerpa (University of California, Merced), Wan Du (University of California Merced) | |
| HyperAlloc: Efficient VM Memory De/Inflation via Hypervisor-Shared Page-Frame Allocators | Lars Wrenger (Leibniz Universität Hannover), Kenny Albes (Leibniz Universität Hannover), Marco Wurps (Leibniz Universität Hannover), Christian Dietrich (Technische Universität Braunschweig), Daniel Lohmann (Leibniz Universität Hannover) | |
| FastIOV: Fast Startup of Passthrough Network I/O Virtualization for Secure Containers | Yunzhuo Liu (Shanghai Jiao Tong University & Alibaba Cloud), Junchen Guo (Alibaba Cloud), Bo Jiang (Shanghai Jiao Tong University), Yang Song (Alibaba Cloud), Pengyu Zhang (Alibaba Cloud), Rong Wen (Alibaba Cloud), Biao Lyu (Zhejiang University & Alibaba Cloud), Shunmin Zhu (Hangzhou Feitian Cloud & Alibaba Cloud), Xinbing Wang (Shanghai Jiao Tong University), Song Yang (Alibaba Group) | |
| Hey Hey, My My, Skewness Is Here to Stay: Challenges and Opportunities in Cloud Block Store Traffic | Haonan Wu (Shanghai Jiao Tong University), Erci Xu (Shanghai Jiao Tong University), Ligang Wang (Alibaba Group), Yuandong Hong (Alibaba Group), Changsheng Niu (Alibaba Group), Bo Shi (Alibaba Group), Lingjun Zhu (Alibaba Group), Jinnian He (Alibaba Group), Dong Wu (Alibaba Group), Weidong Zhang (Alibaba Group), Qiuping Wang (Alibaba Group), Changhong Wang (Alibaba Group), Xinqi Chen (Alibaba Group), Guangtao Xue (Shanghai Jiao Tong University), Yi-Chao Chen (Shanghai Jiao Tong University), Dian Ding (Shanghai Jiao Tong University) | |
| Optimizing Task Scheduling in Cloud VMs with Accurate vCPU Abstraction | Edward Guo (Hofstra University), Weiwei Jia (The University of Rhode Island), Xiaoning Ding (New Jersey Institute of Technology), Jianchen Shan (Hofstra University) |
15:40-16:10 Coffee break & Poster session (Wednesday morning presentations) - Location: Catering area
16:10-17:50
| Session-6.1: ML systems - Location: Mees - Chair: Haibo Chen | Title | Authors |
|---|---|---|
| JABAS: Joint Adaptive Batching and Automatic Scaling for DNN Training on Heterogeneous GPUs | Gyeongchan Yun (UNIST), Junesoo Kang (UNIST), Hyunjoon Jeong (UNIST), Sanghyeon Eom (UNIST), Minsung Jang (Samsung SDS), Young-ri Choi (UNIST) | |
| SpaceFusion: Advanced Deep Learning Operator Fusion via Space-Mapping Graph | Liang Zhu (Shanghai Jiao Tong University), Jianguo Yao (Shanghai Jiao Tong University), Haibing Guan (Shanghai Jiao Tong University) | |
| Groot: Graph-Centric Row Reordering with Tree for Sparse Matrix Multiplications on Tensor Cores | YuAng Chen (The Chinese University of Hong Kong), Jiadong Xie (The Chinese University of Hong Kong), Siyi Teng (The Chinese University of Hong Kong), Wenqi Zeng (Hong Kong University of Science and Technology), Jeffrey Xu Yu (The Chinese University of Hong Kong) | |
| SuperFE: A Scalable and Flexible Feature Extractor for ML-based Traffic Analysis Applications | Menghao Zhang (Beihang University), Guanyu Li (Tsinghua University), Cheng Guo (Tsinghua University), Renyu Yang (Beihang University), Shicheng Wang (Tsinghua University), Han Bao (Tsinghua University), Xiao Li (Tsinghua University), Mingwei Xu (Tsinghua University), Tianyu Wo (Beihang University), Chunming Hu (Beihang University) |
| Session-6.2: Memory Tiering and Disaggregation - Location: Diamond - Chair: Baptiste Lepers | Title | Authors |
|---|---|---|
| Chrono: Meticulous Hotness Measurement and Flexible Page Migration for Memory Tiering | Zhenlin Qi (Shanghai Jiao Tong University), Shengan Zheng (Shanghai Jiao Tong University), Ying Huang (Intel), Yifeng Hui (Shanghai Jiao Tong University), Bowen Zhang (Shanghai Jiao Tong University), Linpeng Huang (Shanghai Jiao Tong University), Hong Mei (Shanghai Jiao Tong University) | |
| PET: Proactive Demotion for Efficient Tiered Memory Management | Wanju Doh (Seoul National University), Yaebin Moon (Samsung Electronics), Seoyoung Ko (Seoul National University), Seunghwan Chung (Seoul National University), Kwanhee Kyung (Seoul National University), Eojin Lee (Inha University), Jung Ho Ahn (Seoul National University) | |
| Adios to Busy-Waiting for Microsecond-scale Memory Disaggregation | Wonsup Yoon (KAIST), Jisu Ok (KAIST), Sue Moon (KAIST), Youngjin Kwon (KAIST) | |
| Deft: A Scalable Tree Index for Disaggregated Memory | Jing Wang (Tsinghua University), Qing Wang (Tsinghua University), Yuhao Zhang (Tsinghua University), Jiwu Shu (Tsinghua University) |
19:00- EuroSys & ASPLOS Banquet - Location: ss Rotterdam
Thursday 3rd April
08:30-9:00- Registration and Welcome Coffee - Location: Catering area
09:00-10:40
| Session-7.1: Cloud and Serverless Computing - Location: Rotterdam hall 1B - Chair: Martin Kleppmann | Title | Authors |
|---|---|---|
| SeBS-Flow: Benchmarking Serverless Cloud Function Workflows | Larissa Schmid (Karlsruhe Institute of Technology), Marcin Copik (ETH Zurich), Alexandru Calotoiu (ETH Zurich), Laurin Brandner (ETH Zurich), Anne Koziolek (Karlsruhe Institute of Technology), Torsten Hoefler (ETH Zurich) | |
| AlloyStack: A Library Operating System for Serverless Workflow Applications | Jianing You (Tianjin University), Kang Chen (Tsinghua University), Laiping Zhao (Tianjin University), Yiming Li (Tianjin University), Yichi Chen (Tianjin University), Yuxuan Du (Tianjin University), Yanjie Wang (Tianjin University), Luhang Wen (Tianjin University), Keyang Hu (Tsinghua University), Keqiu Li (Tianjin University) | |
| Serverless Cold Starts and Where to Find Them | Artjom Joosen (Huawei), Ahmed Hassan (Huawei), Martin Asenov (Huawei), Rajkarn Singh (Huawei), Luke Darlow (Huawei), Jianfeng Wang (Huawei), Qiwen Deng (Huawei), Adam Barker (Huawei, University of St Andrews) | |
| TUNA: Tuning Unstable and Noisy Cloud Applications | Johannes Freischuetz (University of Wisconsin - Madison), Konstantinos Kanellis (University of Wisconsin-Madison), Brian Kroth (Microsoft), Shivaram Venkataraman (University of Wisconsin-Madison) |
| Session-7.2: Storage - Location: Diamond - Chair: Xiaosong Ma | Title | Authors |
|---|---|---|
| Solid State Drive Targeted Memory-Efficient Indexing for Universal I/O Patterns and Fragmentation Degrees | Junsu Im (POSTECH), Jeonggyun Kim (DGIST), Seonggyun Oh (DGIST), Jinhyung Koo (POSTECH), Juhyung Park (DGIST), Hoon Sung Chwa (DGIST), Sam H. Noh (Virginia Tech), Sungjin Lee (POSTECH) | |
| Daredevil: Rescue Your Flash Storage from Inflexible Kernel Storage Stack | Junzhe Li (The University of Hong Kong), Ran Shu (Microsoft Research), Jiayi Lin (The University of Hong Kong), Qingyu Zhang (The University of Hong Kong), Ziyue Yang (Microsoft Research), Jie Zhang (Peking University), Yongqiang Xiong (Microsoft Research), Chenxiong Qian (The University of Hong Kong) | |
| Overcoming the Last Mile between Log-Structured File Systems and Persistent Memory via Scatter Logging | Yifeng Zhang (Harbin Institute of Technology, Shenzhen), Yanqi Pan (Harbin institute of Technology, Shenzhen), Hao Huang (Harbin Institute of Technology, Shenzhen), Yuchen Shan (Harbin Institute of Technology, Shenzhen), Wen Xia (Harbin Institute of Technology, Shenzhen) | |
| Garbage Collection Does Not Only Collect Garbage: Piggybacking-Style Defragmentation for Deduplicated Backup Storage | Dingbang Liu (Harbin Institute of Technology, Shenzhen), Xiangyu Zou (Harbin Institute of Technology, Shenzhen), Tao Lu (DapuStor), Philip Shilane (Dell Technologies), Wen Xia (Harbin Institute of Technology, Shenzhen), Wenxuan Huang (Harbin Institute of Technology, Shenzhen), Yanqi Pan (Harbin Institute of Technology, Shenzhen), Hao Huang (Harbin Institute of Technology, Shenzhen) |
10:40-11:10 Coffee break - Location: Catering area
11:10-12:30
| Session-8.1: Bug finding - Location: Penn - Chair: Julia Lawall | Title | Authors |
|---|---|---|
| Understanding the Linux Kernel, Visually | Hanzhi Liu (Nanjing University), Yanyan Jiang (Nanjing University), Chang Xu (Nanjing University) | |
| Understanding and Detecting SQL Function Bugs: Using Simple Boundary Arguments to Trigger Hundreds of DBMS Bugs | Jingzhou Fu (Tsinghua University), Jie Liang (Tsinghua University), Zhiyong Wu (Tsinghua University), Yanyang Zhao (Tsinghua University), Shanshan Li (National University of Defense Technology), Yu Jiang (Tsinghua University) | |
| BESA: Extending Bugs Triggered by Runtime Testing via Static Analysis | Jia-Ju Bai (Beihang University) | |
| HawkSet: Automatic, Application-Agnostic, and Efficient Concurrent PM Bug Detection | João Oliveira (INESC-ID, IST), João Gonçalves (INESC-ID & IST U. Lisboa), Miguel Matos (IST Lisbon) |
| Session-8.2: Storage - Location: Van Oldebarneveldt - Chair: Yu Hua | Title | Authors |
|---|---|---|
| Heimdall: Optimizing Storage I/O Admission with Extensive Machine Learning Pipeline | Daniar H. Kurniawan (University of Chicago and MangoBoost Inc.), Rani Ayu Putri (Bandung Institute of Technology and University of Chicago), Peiran Qin (University of Chicago), Kahfi S. Zulkifli (Bandung Institute of Technology), Ray A. O. Sinurat (University of Chicago), Janki Bhimani (Florida International University), Sandeep Madireddy (Argonne National Laboratory), Achmad Imam Kistijantoro (Bandung Institute of Technology), Haryadi Gunawi (University of Chicago) | |
| Cheetah: Metadata Aggregation for Fast Object Storage without Distributed Ordering | Yiming Zhang (SJTU & XMU), Li Wang (KylinSoft), Shengyun Liu (SJTU), Shun Gai (NICE XLAB), Haonan Wang (NICE XLAB), Xin Yao (Huawei Theory Lab), Meiling Wang (Huawei Theory Lab), Kai Chen (HKUST), Dongsheng Li (NUDT), Jiwu Shu (Tsinghua University) | |
| Towards Efficient Flash Caches with Emerging NVMe Flexible Data Placement SSDs | Michael Allison (Samsung Electronics), Arun George (Samsung Electronics), Javier Gonzalez (Samsung Electronics), Dan Helmick (Samsung Electronics), Vikash Kumar (Samsung Electronics), Roshan R Nair (Samsung Electronics), Vivek Shah (Samsung Electronics) | |
| Pre-Stores: Proactive Software-guided Movement of Data Down the Memory Hierarchy | Xiaoxiang Wu (The University of Sydney), Baptiste Lepers (Inria and The University of Neuchâtel), Willy Zwaenepoel (University of Sydney) |
12:30-14:00 Lunch - Location: Catering area
14:00-15:40
| Session-9.1: Security and Privacy - Location: Van Oldebarneveldt - Chair: Anjo Vahldiek-Oberwagner | Title | Authors |
|---|---|---|
| RAKIS: Secure Fast I/O Primitives Across Trust Boundaries on Intel SGX | Mansour Alharthi (Georgia Institute of Technology), Fan Sang (Georgia Institute of Technology), Dmitrii Kuvaiskii (Intel Labs), Mona Vij (Intel Labs), Taesoo Kim (Georgia Institute of Technology) | |
| DPack: Efficiency-Oriented Privacy Budget Scheduling | Pierre Tholoniat (Columbia University), Kelly Kostopoulou (Columbia University), Mosharaf Chowdhury (University of Michigan), Asaf Cidon (Columbia University), Roxana Geambasu (Columbia University), Mathias Lécuyer (University of British Columbia), Junfeng Yang (Columbia University) | |
| Erebor: A Drop-In Sandbox Solution for Private Data Processing in Untrusted Confidential Virtual Machines | Chuqi Zhang (National University of Singapore), Rahul Priolkar (Arizona State University), Yuancheng Jiang (National University of Singapore), Yuan Xiao (Intel Labs), Mona Vij (Intel Labs), Zhenkai Liang (National University of Singapore), Adil Ahmad (Arizona State University) | |
| A Hardware-Software Co-Design for Efficient Secure Containers | Jiacheng Shi (Shanghai Jiao Tong University), Yang Yu (Shanghai Jiao Tong University), Jinyu Gu (Shanghai Jiao Tong University), Yubin Xia (Shanghai Jiao Tong University) | |
| Seal: Towards Diverse Specification Inference for Linux Interfaces from Security Patches | Wei Chen (The Hong Kong University of Science and Technology), Bowen Zhang (The Hong Kong University of Science and Technology), Chengpeng Wang (HKUST), Wensheng Tang (The Hong Kong University of Science and Technology), Charles Zhang (HKUST) |
| Session-9.2: LLM Training - Location: Rotterdam hall 1B - Chair: Jacob Gorm Hansen | Title | Authors |
|---|---|---|
| MEPipe: Democratizing LLM Training with Memory-Efficient Slice-Level Pipeline Scheduling on Cost-Effective Accelerators | Zhenbo Sun (Tsinghua University), Shengqi Chen (Tsinghua University), Yuanwei Wang (Tsinghua University), Jian Sha (Tsinghua University), Guanyu Feng (Zhipu AI), Wenguang Chen (Tsinghua University) | |
| HybridFlow: A Flexible and Efficient RLHF Framework | Guangming Sheng (The University of Hong Kong), Chi Zhang (ByteDance), Zilingfeng Ye (ByteDance), Xibin Wu (ByteDance), Wang Zhang (ByteDance), Ru Zhang (ByteDance), Yanghua Peng (ByteDance), Haibin Lin (ByteDance), Chuan Wu (The University of Hong Kong) | |
| Mist: Efficient Distributed Training of Large Language Models via Memory-Parallelism Co-Optimization | Zhanda Zhu (University of Toronto, CentML, Vector Institute), Christina Giannoula (University of Toronto), Muralidhar Andoorveedu (CentML), Qidong Su (University of Toronto, CentML, Vector Institute), Karttikeya Mangalam (UC Berkeley), Bojian Zheng (Independent Researcher), Gennady Pekhimenko (CentML, University of Toronto, Vector Institute) | |
| Hourglass: Enabling Efficient Split Federated Learning with Data Parallelism | Qiang He (Huazhong University of Science and Technology), Kaibin Wang (Swinburne University of Technology), Zeqian Dong (Swinburne University of Technology), Liang Yuan (University of Adelaide), Feifei Chen (Deakin University), Hai Jin (Huazhong University of Science and Technology), Yun Yang (Swinburne University of Technology) | |
| FlowCheck: Decoupling Checkpointing and Training of Large-Scale Models | Zimeng Huang (Shanghai Jiao Tong University & Alibaba Cloud), Hao Nie (Alibaba Cloud & Peking University), Haonan Jia (Alibaba Cloud), Bo Jiang (Shanghai Jiao Tong University), Junchen Guo (Alibaba Cloud), Jianyuan Lu (Alibaba Cloud), Rong Wen (Alibaba Cloud), Biao Lyu (Zhejiang University & Alibaba Cloud), Shunmin Zhu (Hangzhou Feitian Cloud & Alibaba Cloud), Xinbing Wang (Shanghai Jiao Tong University) |
15:40-16:10 Coffee break - Location: Catering area
16:10-17:30
| Session-10.1: Network - Van Oldebarneveldt - Chair: Rong Chen | Title | Authors |
|---|---|---|
| Atlas: Towards Real-Time Verification in Large-Scale Networks via a Native Distributed Architecture | Mingxiao Ma (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Yuehan Zhang (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Jingyu Wang (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Bo He (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Chenyang Zhao (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Qi Qi (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Zirui Zhuang (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Haifeng Sun (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Lingqi Guo (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Yuebin Guo (State Key Laboratory of Networking and Switching Technology, Beijing University of Post and Telecommunication), Gong Zhang (Huawei Technologies), Jianxin Liao (State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications) | |
| Occamy: A Preemptive Buffer Management for On-chip Shared-memory Switches | Danfeng Shan (Xi'an Jiaotong University), Yunguang Li (Xi'an Jiaotong University), Jinchao Ma (Xi'an Jiaotong University), Zhenxing Zhang (Huawei), Zeyu Liang (Xi'an Jiaotong University), Xinyu Wen (Xi'an Jiaotong University), Hao Li (Xi'an Jiaotong University), Wanchun Jiang (Central South University), Nan Li (Huawei), Fengyuan Ren (Tsinghua University) | |
| Phantom: Virtualizing Switch Register Resources for Accurate Sketch-based Network Measurement | Xiang Chen (Zhejiang University), Hongyan Liu (Zhejiang University), Zhengyan Zhou (Zhejiang University), Xi Sun (Zhejiang University), Wenbin Zhang (Zhejiang University), Hongyang Du (The University of Hong Kong), Dong Zhang (Fuzhou University), Xuan Liu (Yangzhou University and Southeast University), Haifeng Zhou (Zhejiang University), Dusit Niyato (Nanyang Technological University), Qun Huang (Peking University), Chunming Wu (Zhejiang University), Kui Ren (Zhejiang University) |
| Session-10.2: Cloud and Networking - Rotterdam hall 1B - Chair: Laiping Zhao | Title | Authors |
|---|---|---|
| Eva: Cost-Efficient Cloud-Based Cluster Scheduling | Tzu-Tao Chang (University of Wisconsin-Madison), Shivaram Venkataraman (University of Wisconsin-Madison) | |
| Byte vSwitch: A High-Performance Virtual Switch for Cloud Networking | Xin Wang (ByteDance Inc.), Deguo Li (ByteDance Inc.), Zhihong Wang (ByteDance Inc.), Lidong Jiang (ByteDance Inc.), Shubo Wen (ByteDance Inc.), Daxiang Kang (ByteDance Inc.), Engin Arslan (ByteDance Inc.), Peng He (ByteDance Inc.), Xinyu Qian (ByteDance Inc.), Bin Niu (ByteDance Inc.), Jianwen Pi (ByteDance Inc.), Xiaoning Ding (ByteDance Inc.), Ke Lin (ByteDance Inc.), Hao Luo (ByteDance Inc.) |