venue: OSDI
year: 2025
papers:
- title: 'Basilisk: Using Provenance Invariants to Automate Proofs of Undecidable Protocols'
  authors:
  - Tony Nuda Zhang
  - Keshav Singh
  - Tej Chajed
  - Manos Kapritsos
  - Bryan Parno
  reason: "Automates the construction of correctness proofs for distributed protocols that were previously considered undecidable, advancing the state of the art in verified systems."
- title: 'Mako: Speculative Distributed Transactions with Geo-Replication'
  authors:
  - Weihai Shen
  - Yang Cui
  - Siddhartha Sen 0001
  - Sebastian Angel
  - Shuai Mu 0001
  reason: "Combines speculative execution with geo-replication to deliver low-latency distributed transactions without sacrificing consistency, addressing a fundamental tension in wide-area systems."
- title: Low End-to-End Latency atop a Speculative Shared Log with Fix-Ante Ordering
  authors:
  - Shreesha G. Bhat
  - Tony Hong
  - Xuhao Luo
  - Jiyu Hu
  - Aishwarya Ganesan
  - Ramnatthan Alagappan
  reason: "Introduces fix-ante ordering to achieve low latency on a shared log without sacrificing throughput, offering a new design point for log-based distributed storage."
- title: 'Okapi: Decoupling Data Striping and Redundancy Grouping in Cluster File Systems'
  authors:
  - Sanjith Athlur
  - Timothy Kim
  - Saurabh Kadekodi
  - Francisco Maturana
  - Xavier Ramos
  - Arif Merchant
  - K. V. Rashmi
  - Gregory R. Ganger
  reason: "Challenges a long-standing coupling in erasure-coded cluster file systems, enabling independent optimization of striping and redundancy with measurable gains in production workloads."
- title: 'PoWER Never Corrupts: Tool-Agnostic Verification of Crash Consistency and Corruption Detection'
  authors:
  - Hayley LeBlanc
  - Jacob R. Lorch
  - Chris Hawblitzel
  - Cheng Huang
  - Yiheng Tao
  - Nickolai Zeldovich
  - Vijay Chidambaram
  reason: "Provides a tool-agnostic framework for formally verifying crash consistency and corruption detection in storage systems, raising the bar for storage software correctness."
- title: 'EMT: An OS Framework for New Memory Translation Architectures'
  authors:
  - Siyuan Chai 0001
  - Jiyuan Zhang 0003
  - Jongyul Kim 0001
  - Alan Wang
  - Fan Chung
  - Jovan Stojkovic
  - Weiwei Jia 0001
  - Dimitrios Skarlatos 0002
  - Josep Torrellas
  - Tianyin Xu
  reason: "Defines an OS abstraction layer that decouples applications from hardware-specific memory translation mechanisms, enabling future memory architectures to be adopted without OS rewrites."
- title: 'XSched: Preemptive Scheduling for Diverse XPUs'
  authors:
  - Weihang Shen
  - Mingcong Han
  - Jialong Liu
  - Rong Chen 0001
  - Haibo Chen 0001
  reason: "Generalises preemptive scheduling to heterogeneous accelerators (XPUs), providing a unified OS-level mechanism for fair and responsive multi-tenant accelerator sharing."
- title: Extending Applications Safely and Efficiently
  authors:
  - Yusheng Zheng
  - Tong Yu
  - Yiwei Yang 0002
  - Yanpeng Hu
  - Xiaozheng Lai
  - Dan Williams 0001
  - Andi Quinn
  reason: "Presents a principled model for safe, efficient application extensibility that generalises beyond eBPF, with implications for the design of future OS extension mechanisms."
- title: 'NanoFlow: Towards Optimal Large Language Model Serving Throughput'
  authors:
  - Kan Zhu
  - Yufei Gao
  - Yilong Zhao 0002
  - Liangyu Zhao
  - Gefei Zuo
  - Yile Gu
  - Dedong Xie
  - Zihao Ye 0001
  - Keisuke Kamahori
  - Chien-Yu Lin
  - Ziren Wang
  - Stephanie Wang
  - Arvind Krishnamurthy
  - Baris Kasikci
  reason: "Analytically characterises the throughput ceiling for LLM serving and proposes a system that approaches that bound through fine-grained intra-device parallelism."
- title: 'WaferLLM: Large Language Model Inference at Wafer Scale'
  authors:
  - Congjie He
  - Yeqi Huang
  - Pei Mu 0003
  - Ziming Miao
  - Jilong Xue
  - Lingxiao Ma
  - Fan Yang 0024
  - Luo Mai
  reason: "Demonstrates end-to-end LLM inference on wafer-scale hardware, tackling novel challenges in memory, communication, and fault tolerance at an unprecedented scale of integration."
- title: 'Mirage: A Multi-Level Superoptimizer for Tensor Programs'
  authors:
  - Mengdi Wu
  - Xinhao Cheng
  - Shengyu Liu
  - Chunan Shi
  - Jianan Ji
  - Man Kit Ao
  - Praveen Velliengiri
  - Xupeng Miao
  - Oded Padon
  - Zhihao Jia
  reason: "Extends tensor program superoptimisation to multiple abstraction levels, discovering non-obvious kernel fusions that outperform hand-tuned implementations for ML workloads."
- title: 'Training with Confidence: Catching Silent Errors in Deep Learning Training with Automated Proactive Checks'
  authors:
  - Yuxuan Jiang 0016
  - Ziming Zhou
  - Boyu Xu 0005
  - Beijie Liu
  - Runhui Xu
  - Peng Huang 0005
  reason: "Addresses the underappreciated problem of silent hardware and software errors in large-scale DL training, providing automated proactive checks that catch failures before they corrupt long training runs."
- title: 'Compass: Encrypted Semantic Search with High Accuracy'
  authors:
  - Jinhao Zhu
  - Liana Patel
  - Matei Zaharia
  - Raluca Ada Popa
  reason: "Enables accurate semantic (vector) search over encrypted data, bridging the gap between privacy-preserving computation and modern retrieval workloads in cloud-hosted RAG systems."