CrossFS: A Cross-layered Direct-Access File System
Yujie Ren, Changwoo Min, Sudarsun Kannan
Abstract
We design CrossFS, a cross-layered direct-access file system disaggregated across user-level, firmware, and kernel layers for scaling I/O performance and improving concurrency. CrossFS is designed to exploit host-and device-level compute capabilities. For concurrency with or without data sharing across threads and processes, CrossFS introduces a file descriptor-based concurrency control that maps each file descriptor to one hardware-level I/O queue. This design allows CrossFS's firmware component to process disjoint access across file descriptors concurrently. CrossFS delegates concurrency control to powerful host-CPUs, which convert the file descriptor synchronization problem into an I/O queue request ordering problem. To guarantee crash consistency in the cross-layered design, CrossFS exploits byte-addressable nonvolatile memory for I/O queue persistence and designs a lightweight firmware-level journaling mechanism. Finally, CrossFS designs a firmware-level I/O scheduler for efficient dispatch of file descriptor requests. Evaluation of emulated CrossFS on storage-class memory shows up to 4.87X concurrent access gains for benchmarks and 2.32X gains for real-world applications over the state-of-the-art kernel, userlevel, and firmware file systems.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext e2ed1bb8-5036-4c04-960b-942a6f5d48f7Cited by top-tier papers16
- GPU-Initiated On-Demand High-Throughput Storage Access in the BaM System ArchitectureZaid Qureshi, Vikram Sharma Mailthody, Isaac Gelado, Seungwon Min et al.ASPLOS 2023 · 48 citations
- ODINFS: Scaling PM Performance with Opportunistic DelegationDiyu Zhou, Yuchen Qian, Vishal Gupta, Zhifei Yang et al.OSDI 2022 · 29 citations
- MT^2: Memory Bandwidth Regulation on Hybrid NVM/DRAM PlatformsJifei Yi, Benchao Dong, Mingkai Dong, Ruizhe Tong et al.FAST 2022 · 23 citations
- OmniCache: Collaborative Caching for Near-storage AcceleratorsJian Zhang, Yujie Ren, Marie Nguyen, Changwoo Min et al.FAST 2024 · 18 citations
- FusionFS: Fusing I/O Operations using CISCOps in Firmware File SystemsJian Zhang, Yujie Ren, Sudarsun KannanFAST 2022 · 17 citations
Builds on6
- Spectre Attacks: Exploiting Speculative ExecutionPaul Kocher, Jann Horn, Anders Fogh, Daniel Genkin et al.S&P 2019 · 2,435 citations
- Meltdown: Reading Kernel Memory from User SpaceMoritz Lipp, Michael Schwarz, Daniel Gruss, Thomas Prescher et al.USENIX Security 2018 · 1,456 citations
- FPGA-Accelerated Compactions for LSM-based Key-Value StoreTeng Zhang, Jianying Wang, Xuntao Cheng, Hao Xu et al.FAST 2020 · 99 citations
- Durable Transactional Memory Can Scale with TimestoneMadhava Krishnan Ramanathan, Jaeho Kim, Ajit Mathew, Xinwei Fu et al.ASPLOS 2020 · 46 citations
- Avoiding scheduler subversion using scheduler-cooperative locksYuvraj Patel, Leon Yang, Leo Prasath Arulraj, Andrea C. Arpaci-Dusseau et al.EuroSys 2020 · 6 citations
Related papers
- Scalable Persistent Memory File System with Kernel-Userspace CollaborationYoumin Chen, Youyou Lu, Bohong Zhu, Andrea C. Arpaci-Dusseau et al.FAST 2021 · 15 citations
- MadFS: Per-File Virtualization for Userspace Persistent Memory FilesystemsShawn Zhong, Chenhao Ye, Guanzhou Hu, Suyan Qu et al.FAST 2023 · 27 citations
- Scale and Performance in a Filesystem Semi-MicrokernelJing Liu, Anthony Rebello, Yifan Dai, Chenhao Ye et al.SOSP 2021 · 14 citations
- Exploring the Asynchrony of Slow Memory Filesystem with EasyIOBohong Zhu, Youmin Chen, Jiwu ShuEuroSys 2024 · 4 citations
- ByteFS: System Support for (CXL-based) Memory-Semantic Solid-State DrivesShaobo Li, Yirui Eric Zhou, Hao Ren, Jian HuangASPLOS 2025 · 13 citations
