A Flattened Metadata Service for Distributed File Systems

被引:8
作者
Li, Siyang [1 ]
Liu, Fenlin [1 ]
Shu, Jiwu [2 ]
Lu, Youyou [2 ]
Li, Tao [3 ]
Hu, Yang [3 ]
机构
[1] State Key Lab Math Engn & Adv Comp, Zhengzhou 450002, Henan, Peoples R China
[2] Tsinghua Univ, Beijing 100084, Peoples R China
[3] Univ Florida, Gainesville, FL 32611 USA
基金
国家重点研发计划; 中国国家自然科学基金;
关键词
Metadata server; distributed file system; storage system;
D O I
10.1109/TPDS.2018.2842183
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Key-Value stores provide scalable metadata service for distributed file systems. However, the metadata's organization itself, which is organized using a directory tree structure, does not fit the key-value access pattern, thereby limiting the performance. To address this issues, we propose a distributed file system with a flattened and fine-grained division metadata service, LocoMeta, to bridge the performance gap between file system metadata and key-value stores. LocoMeta is designed to bridge the gap between file metadata to key-value store with two techniques. First, LocoMeta flattens the directory content and structure, which organizes file and directory index nodes in a flat space while reversely indexing the directory entries. Second, it exploits a fine-grained division method to improve the key-value access performance. Evaluations show that LocoMeta with eight nodes boosts the metadata throughput by five times, which approaches 93 percent throughput of a single-node key-value store, compared to 18 percent in the state-of-the-art IndexFS.
引用
收藏
页码:2641 / 2657
页数:17
相关论文
共 50 条
  • [31] Web services usage in distributed file systems
    Pais, V. F.
    FUSION ENGINEERING AND DESIGN, 2010, 85 (3-4) : 419 - 422
  • [32] Modeling of Distributed File Systems for Practical Performance Analysis
    Wu, Yongwei
    Ye, Feng
    Chen, Kang
    Zheng, Weimin
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2014, 25 (01) : 156 - 166
  • [33] Comparison of Spark Resource Managers and Distributed File Systems
    Salehian, Solmaz
    Yan, Yonghong
    PROCEEDINGS OF 2016 IEEE INTERNATIONAL CONFERENCES ON BIG DATA AND CLOUD COMPUTING (BDCLOUD 2016) SOCIAL COMPUTING AND NETWORKING (SOCIALCOM 2016) SUSTAINABLE COMPUTING AND COMMUNICATIONS (SUSTAINCOM 2016) (BDCLOUD-SOCIALCOM-SUSTAINCOM 2016), 2016, : 567 - 572
  • [34] Adaptive metadata rebalance in exascale file system
    Cha, Myung-Hoon
    Kim, Dong-Oh
    Kim, Hong-Yeon
    Kim, Young-Kyun
    JOURNAL OF SUPERCOMPUTING, 2017, 73 (04) : 1337 - 1359
  • [35] Adaptive metadata rebalance in exascale file system
    Myung-Hoon Cha
    Dong-Oh Kim
    Hong-Yeon Kim
    Young-Kyun Kim
    The Journal of Supercomputing, 2017, 73 : 1337 - 1359
  • [36] Atomic accesses to replicated files in reliable distributed file systems
    Jia, XH
    Shimizu, K
    Maekawa, M
    COMPUTER SYSTEMS SCIENCE AND ENGINEERING, 1996, 11 (03): : 151 - 158
  • [37] A NOVEL APPROACH FOR REPLICA SYNCHRONIZATION IN HADOOP DISTRIBUTED FILE SYSTEMS
    Vini, Miss. J.
    Nallathamby, Rachel
    Robin, C. R. Rene
    BIG DATA, CLOUD AND COMPUTING CHALLENGES, 2015, 50 : 590 - 595
  • [38] Impact of Metadata Server on a Large Scale File System
    Patgiri, Ripon
    Nayak, Sabuzima
    Borgohain, Samir Kumar
    2018 IEEE COLOMBIAN CONFERENCE ON COMMUNICATIONS AND COMPUTING (COLCOM), 2018,
  • [39] The Trellis security infrastructure for overlay metacomputers and bridged distributed file systems
    Lu, Paul
    Closson, Michael
    Macdonell, Cam
    Nalos, Paul
    Ngo, Danny
    Kan, Morgan
    Lee, Mark
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2006, 66 (09) : 1181 - 1188
  • [40] Block I/O Scheduling on Storage Servers of Distributed File Systems
    Liao, Jianwei
    Yin, Dong
    Peng, Xiaoning
    JOURNAL OF GRID COMPUTING, 2018, 16 (02) : 299 - 316