UFCR: An efficient I/O method for parallel file system

被引：0
|
作者：
Huo, Yanmei [1 ]
Ju, Jiubin [1 ]
Hu, Liang [1 ]
机构：
[1] Jilin Univ, Coll Comp Sci & Technol, Changsha 130012, Peoples R China
来源：
SEVENTH INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED COMPUTING, APPLICATIONS AND TECHNOLOGIES, PROCEEDINGS | 2006年
基金：
中国国家自然科学基金;
关键词：
D O I：
暂无
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
One important way of solving 110 bottleneck is to develop parallel file system. Research showed that the strided file access pattern in scientific applications combined with current file system methods to perform these accesses lead to unacceptable performance for large data access. To enhance performance of discontiguous data access, we have created a new high performance 110 method: the user-defined file view and combining 110 requests (UFCR). We have implemented our ideas on WPFS. Experimentation results showed that our method can substantially enhance the performance of scientific applications.
引用
收藏
页码：223 / +
页数：3
相关论文
共 50 条
  • [1] Benchmarking Parallel File System Sensitiveness to I/O patterns
    Chasapis, Konstantinos
    Vet, Jean-Yves
    Acquaviva, Jean-Thomas
    [J]. 2019 IEEE 27TH INTERNATIONAL SYMPOSIUM ON MODELING, ANALYSIS, AND SIMULATION OF COMPUTER AND TELECOMMUNICATION SYSTEMS (MASCOTS 2019), 2019, : 427 - 428
  • [2] Distributed parallel file system for I/O intensive parallel computing on clusters
    Domínguez-Domínguez, S
    Buenabad-Chávez, J
    [J]. 2004 1st International Conference on Electrical and Electronics Engineering (ICEEE), 2004, : 194 - 199
  • [3] Enabling dynamic file I/O path selection at runtime for parallel file system
    Xiuqiao Li
    Limin Xiao
    Meikang Qiu
    Bin Dong
    Li Ruan
    [J]. The Journal of Supercomputing, 2014, 68 : 996 - 1021
  • [4] Enabling dynamic file I/O path selection at runtime for parallel file system
    Li, Xiuqiao
    Xiao, Limin
    Qiu, Meikang
    Dong, Bin
    Ruan, Li
    [J]. JOURNAL OF SUPERCOMPUTING, 2014, 68 (02): : 996 - 1021
  • [5] FILE CONCEPTS FOR PARALLEL I/O
    CROCKETT, TW
    [J]. PROCEEDINGS : SUPERCOMPUTING 89, 1989, : 574 - 579
  • [6] A File Assignment Strategy for Parallel I/O System with Minimum I/O Contention Probability
    Dong, Bin
    Li, Xiuqiao
    Xiao, Limin
    Ruan, Li
    [J]. GRID AND DISTRIBUTED COMPUTING, 2011, 261 : 445 - 454
  • [7] Parallel I/O Architecture Modelling Based on File System Counters
    El Sayed, Salem
    Bolten, Matthias
    Pleiter, Dirk
    [J]. HIGH PERFORMANCE COMPUTING, ISC HIGH PERFORMANCE 2016 INTERNATIONAL WORKSHOPS, 2016, 9945 : 627 - 637
  • [8] Using file system counters in modelling parallel I/O architectures
    El Sayed, Salem
    Bolten, Matthias
    Pleiter, Dirk
    [J]. OPERATING SYSTEMS REVIEW, 2016, 50 (02) : 37 - 46
  • [9] MPI-I/O with a shared file pointer using a parallel virtual file system in remote I/O operations
    Tsujita, Y
    [J]. DISTRIBUTED AND PARALLEL SYSTEMS: CLUSTER AND GRID COMPUTING, 2005, 777 : 47 - 54
  • [10] Parallel File System Analysis Through Application I/O Tracing
    Wright, S. A.
    Hammond, S. D.
    Pennycook, S. J.
    Bird, R. F.
    Herdman, J. A.
    Miller, I.
    Vadgama, A.
    Bhalerao, A.
    Jarvis, S. A.
    [J]. COMPUTER JOURNAL, 2013, 56 (02): : 141 - 155
12345