期刊文献+

基于可分组设计的部分重复码研究 被引量:5

Research on fractional repetition codes based on group divisible designs
下载PDF
导出
摘要 针对最小带宽再生情形下的有效修复问题,提出了一种新型部分重复(FR,fractional repetition)码设计。该设计由外部最大距离可分(MDS,maximum distance separable)码和内部重复码组成,称为GDDBFR(group divisible design based FR)码,可以达到随机访问模式下的系统存储容量,并且能够在很大范围内选择构造参数。理论分析指出,尽管GDDBFR码采用基于表格的修复方式,但通常具有大量的节点修复选择方案。此外,实验结果表明,与传统的RS(Reed-Solomon)码和再生码相比,GDDBFR码可以显著地减少失效修复时间。 A novel design of FR(fractional repetition) codes was proposed which aims at providing efficient repair at the minimum bandwidth regenerating point. The design consisted of an outer MDS(maximum distance separable) code and an inner repetition code, called GDDBFR(group divisible design based FR) codes. The proposed codes can achieve the system storage capacity under the random access model and are available for a wide range of parameters. Despite of the table-based repair, theoretical analysis identifies that GDDBFR codes generally have large node repair alternatives. Furthermore, experimental results show that GDDBFR codes can significantly reduce the failure repair time when compared with legacy RS(Reed-Solomon) codes and regenerating codes in the domain.
出处 《通信学报》 EI CSCD 北大核心 2015年第2期98-105,共8页 Journal on Communications
基金 国家重点基础研究计划("973计划")基金资助项目(2012CB315904) 国家自然科学基金资助项目(61179028) 广东省自然科学基金资助项目(S2013020012822) 深圳市基础研究基金资助项目(JCYJ20140417144423192 JCYJ20130331144502026)~~
关键词 部分重复码 可分组设计 存储容量 节点修复选择度 修复时间 fractional repetition codes group divisible designs storage capacity node repair alternativity repair time
  • 相关文献

参考文献13

  • 1GHEMAWAT S, GOBIOFF H, LEUNG S. The Google file system[A].The 19th ACM Symposium on Operating Systems PrincipIesfC]. LakeGeorge’ New York, USA’ 2003. 29-43.
  • 2DIMAKIS A Q RAMCHANDRAN K, WU Y, et al. A survey onnetwork codes for distributed storage[J]. Proceedings of the IEEE,2011,99(3): 476-489.
  • 3STEVEN J,MURDOCH, PIOTR ZIELINSKI. Sampled traffic analy-sis by intemet-exchange-level adversaries[A]. Proceedings of the 7thInternational Conference on Privacy Enhancing Technologies[C].2007.167-183.
  • 4DIMAKIS A Q GODFREY P B, WU Y, et al. Network coding fordistributed storage systems[J]. IEEE Transactions on InformationTheory, 2010, 56(9): 4539-4551.
  • 5RASHMIK V, SHAH N B’ KUMAR P V. Optimal exact-regeneratingcodes for distributed storage at the MSR and MBR points via a prod-uct-matrix construction[J], IEEE Transactions on Information Theory,2011,57(8): 5227-5239.
  • 6RASHMI K V,SHAH N B, KUMAR P V,et al. Explicit constructionof optimal exact regenerating codes for distributed storage[A]. The47th Annual Allerton Conference on Communication, Control, andComputing[C]. Monticello,IL, USA,2009.1243-1249.
  • 7RASHMI K V,SHAH N B,KUMAR P V,et al Explicit and optimalexact-regenerating codes for the minimum-bandwidth point in distrib-uted storagefA]. IEEE International Symposium on Information Tlie-ory Proceedings[C]. Austin, TX, USA, 2010. 1938-1942.
  • 8VENKATESAN V. Fast Rebuilds in Distributed Storage SystemsUsing Network Coding[R]. Zurich Research Laboratory, IBM Re-search GmbH, Zurich, 2009.
  • 9ROUAYHEB S E,RAMCHANDRAN K. Fractional repetition codesfor repair in distributed storage systems[A]. The 48th Annual AllertonConference on Communication, Control, and ComputingfC]. Allerton,IL,USA, 2010. 1510-1517.
  • 10PAWAR S, NOORSHAMS N, ROUAYHEB S E, et al. Dress codesfor the storage cloud: simple randomized constructions[A]. IEEE In-ternational Symposium on Information Theory Proceedings[C]. St.Petersburg, Russia, 2011. 2338-2342.

同被引文献16

引证文献5

二级引证文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部