分布式存储系统中基于Pyramid码的局部性修复编码
CSTR:
作者:
作者单位:

1. 长安大学信息工程学院西安710064;2. 湖南科技大学计算机科学与工程学院湘潭411201;3. 西安通信学院西安710106

作者简介:

通讯作者:

中图分类号:

TP911.2

基金项目:

国家自然科学基金 (61640006, 61572188)、陕西省自然科学基金 (2016JQ6011) 西安市科技计划项目(2017088CG/RC051(CADX002))、中央高校基本科研业务费专项资金(310850160317)资助项目


Locally repairable codes based on Pyramid codes in distributed storage systems
Author:
Affiliation:

1. School of Information Engineering, Chang’an University, Xi’an 710064, China;2. School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411201, China; 3. Xi’an Communication College, Xi’an 710106, China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为了提高分布式存储系统的存储可靠性以及故障节点的修复效率,提出一种基于Pyramid码的局部性修复编码方案。该编码方案采用Pyramid码的最小可实现编码结构,划分局部修复组,确保较低的修复局部性并实现故障节点的快速修复。性能分析表明,基于Pyramid码的局部性修复编码可实现存储系统中多个故障节点的快速修复,具有较低的修复局部性,且相对于三副本复制策略以及简单再生码,基于Pyramid码的局部性修复编码在存储开销和修复带宽开销方面的性能更优。

    Abstract:

    In order to improve the reliability of distributed storage systems and repair efficiency of failed nodes, locally repairable codes based on Pyramid codes are proposed in this paper. Specifically, adopting the minimum achievable encoding structure of Pyramid codes, the proposed locally repairable codes divide the nodes of distributed storage systems into multiple local repair groups, to achieve lower repair locality and rapid repair of failed nodes. The performance analysis shows that, the proposed locally repairable codes based on Pyramid codes can achieve rapid repair of multiple failed nodes in distributed storage systems, has lower repair locality. Moreover, compared with three copy replication strategies and simple regenerating codes, the proposed locally repairable codes based on Pyramid codes have advantages in the performances of storage overhead and repair bandwidth overhead.

    参考文献
    相似文献
    引证文献
引用本文

王静,张崇,梁伟,刘向阳.分布式存储系统中基于Pyramid码的局部性修复编码[J].电子测量与仪器学报,2017,31(9):1481-1487

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2017-11-06
  • 出版日期:
文章二维码