首页 | 本学科首页   官方微博 | 高级检索  
     

云存储系统中基于MSR码的局部性修复编码
引用本文:王静,梁伟,刘向阳,杨洋.云存储系统中基于MSR码的局部性修复编码[J].北京邮电大学学报,2016,39(4):60-66.
作者姓名:王静  梁伟  刘向阳  杨洋
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, 湖南 湘潭 411201;
3. 西安通信学院 信息服务系, 西安 710106
基金项目:国家自然科学基金项目(61040005;61572188),陕西省自然科学基金项目(2015JM6307),大学生创新创业训练计划项目(201510710131)
摘    要:为了提高分布式云存储系统的存储可靠性和故障节点修复效率,提出一种基于最小存储再生码的局部性修复编码方案. 具体地,构造适用于云存储的系统最小存储再生码,以此码为局部码构造局部性修复编码,确保最大距离可分性质和简单修复特性. 性能分析和仿真结果表明,该局部性修复编码方案可实现云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,相对于三副本复制方式和简单再生码,该局部性修复编码方案在存储开销和修复带宽开销方面的性能更优.

关 键 词:分布式云存储  最小存储再生码  局部性修复编码  
收稿时间:2015-09-25

Locally Repairable Codes Based on MSR Codes in Cloud Storage System
WANG Jing,LIANG Wei,LIU Xiang-yang,YANG Yang.Locally Repairable Codes Based on MSR Codes in Cloud Storage System[J].Journal of Beijing University of Posts and Telecommunications,2016,39(4):60-66.
Authors:WANG Jing  LIANG Wei  LIU Xiang-yang  YANG Yang
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, Hunan Xiangtan 411201, China;
3. School of Information Service, Xi'an Communication College, Xi'an 710106, China
Abstract:In order to improve the reliability of distributed cloud storage system and the efficiency for re-pairing the failed nodes, a scheme of locally repairable coding based on minimum storage regenerating ( MSR) codes was proposed. Specifically, the systematic MSR codes suitable for cloud storage and the locally repairable codes with MSR codes as local codes were constructed respectively to ensure the maxi-mum distance separable ( MDS) property and simple repairing characteristics. Performance analysis and simulation show that the locally repairable coding scheme can realize fast repairing of multiple failed nodes in distributed cloud storage system, and has lower repair locality. Compared with three-copy mode and simple regenerating codes, the locally repairable coding scheme has good performances in storage o-verhead and repair bandwidth overhead.
Keywords:distributed cloud storage  minimum storage regenerating codes  locally repairable codes
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《北京邮电大学学报》浏览原始摘要信息
点击此处可从《北京邮电大学学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号