USENIX ATC2023顶会
P2CACHE: Exploring Tiered Memory for In-Kernel File Systems Caching
Zhen Lin, Lingfeng Xiang, Jia Rao, Hui Lu
摘要
Fast, byte-addressable persistent memory (PM) is becoming a reality in products. However, porting legacy kernel file systems to fully support PM requires substantial effort and encounters the challenge of bridging the gap between blockbased access granularity and byte-addressability. Moreover, new PM-specific file systems remain far from productionready, preventing them from being widely used. In this paper, we propose P 2 CACHE, a novel in-kernel caching mechanism to explore how legacy kernel file systems can effectively evolve in the face of fast, byte-addressable PM. P 2 CACHE exploits a read/write-distinguishable memory hierarchy upon a tiered memory system involving both PM and DRAM. P 2 CACHE leverages PM to serve all write requests for instant data durability and strong crash consistency while using DRAM to serve most read I/Os for high I/O performance. Further, P 2 CACHE employs a simple yet effective synchronization model between PM and DRAM by leveraging devicelevel parallelism. Our evaluation shows that P 2 CACHE can significantly increase the performance of legacy kernel file systems -e.g., by 200x for RocksDB on Ext4 -meanwhile equipping them with instant data durability and strong crash consistency, similar to PM-specialized file systems.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper6
- Nomad: Non-Exclusive Memory Tiering via Transactional Page MigrationLingfeng Xiang, Zhen Lin, Weishu Deng, Hui Lu 等OSDI 2024 · 被引用 59 次
- PolyStore: Exploiting Combined Capabilities of Heterogeneous StorageYujie Ren, David Domingo, Jian Zhang, Paul John 等FAST 2025 · 被引用 6 次
- Boosting File Systems Elegantly: A Transparent NVM Write-ahead Log for Disk File SystemsGuoyu Wang, Xilong Che, Haoyang Wei, Shuo Chen 等FAST 2025 · 被引用 5 次
- Rethinking the Request-to-IO Transformation Process of File Systems for Full Utilization of High-Bandwidth SSDsYekang Zhan, Haichuan Hu, Xiangrui Yang, Qiang Cao 等FAST 2025 · 被引用 4 次
- Keigo: Co-designing Log-Structured Merge Key-Value Stores with a Non-Volatile, Concurrency-aware Storage HierarchyRúben Adão, Zhongjie Wu, Changjun Zhou, Oana Balmau 等VLDB 2025
它引用的顶会 Paper5
- An Empirical Guide to the Behavior and Use of Scalable Persistent MemoryJian Yang, Juno Kim, Morteza Hoseinzadeh, Joseph Izraelevitz 等FAST 2020 · 被引用 470 次
- XRP: In-Kernel Storage Functions with eBPFYuhong Zhong, Haoyu Li, Yu Jian Wu, Ioannis Zarkadas 等OSDI 2022 · 被引用 100 次
- The Storage Hierarchy is Not a Hierarchy: Optimizing Caching on Modern Storage Devices with OrthusKan Wu, Zhihan Guo, Guanzhou Hu, Kaiwei Tu 等FAST 2021 · 被引用 73 次
- Scale and Performance in a Filesystem Semi-MicrokernelJing Liu, Anthony Rebello, Yifan Dai, Chenhao Ye 等SOSP 2021 · 被引用 14 次
- First Responder: Persistent Memory Simultaneously as High Performance Buffer Cache and StorageHyunsub Song, Shean Kim, J. Hyun Kim, Ethan J. H. Park 等USENIX ATC 2021 · 被引用 14 次
相关 Paper
- ctFS: Replacing File Indexing with Hardware Memory Translation through Contiguous File Allocation for Persistent MemoryRuibin Li, Xiang Ren, Xu Zhao, Siwei He 等FAST 2022 · 被引用 43 次
- TL4x: Buffered Durable Transactions on Disk as Fast as in MemoryGal Assa, Andreia Correia, Pedro Ramalhete, Valerio Schiavoni 等PPoPP 2023 · 被引用 8 次
- DaxVM: Stressing the Limits of Memory as a File InterfaceChloe Alverti, Vasileios Karakostas, Nikhita Kunati, Georgios I. Goumas 等MICRO 2022 · 被引用 9 次
- Exploiting Persistent CPU Cache for Scalable Persistent Hash IndexBowen Zhang, Shengan Zheng, Liangxu Nie, Zhenlin Qi 等ICDE 2024 · 被引用 3 次
- HUNTER: Releasing Persistent Memory Write Performance with A Novel PM-DRAM Collaboration ArchitectureYanqi Pan, Yifeng Zhang, Wen Xia, Xiangyu Zou 等DAC 2023 · 被引用 1 次
