期刊文献+

通过复制方式降低修复带宽的新再生码

New regenerating codes reducing repair bandwidth by replication
在线阅读 下载PDF
导出
摘要 再生码通过允许节点传送所存数据的线性组合并增加修复入度,显著地降低了修复带宽,但是增加了参与节点数和磁盘I/O。针对这一缺点,首先通过分析再生码的阈值函数得到一个定理,指出了降低阈值函数的一个充要条件;然后根据该定理构造出了结合复制方式的新再生码。新再生码不增加修复入度也能降低阈值函数,与再生码类似,也有两个特殊点——最小存储量点和最小修复带宽点。通过对两个特殊点的定性与定量分析,发现新再生码可以利用复制来降低单节点修复带宽,并且平均修复带宽和平均磁盘I/O也会减少。 By allowing storage nodes to send linear combinations of their data to the newcomer and increasing repair degree, regenerating codes significantly reduce repair bandwidth overhead. But,they sacrifice repair locality and disk I/O. Against this disadvantage,this paper first gave a theorem which indicated the necessary and sufficient condition to reduce the threshold function in regenerating codes by analyzing the structure of threshold function, then introduced new regenerating codes combined with rephcation based on the theorem. New regenerating codes can reduce threshold function without increasing repair degree, and have also 2 special points-minimum-storage point and minimum-bandwidth point. After qualitative and quantitative analysis, it shows that new regenerating codes can reduce repair bandwidth overhead of per node by replication, and reduce mean repair bandwidth and disk I/O.
出处 《计算机应用研究》 CSCD 北大核心 2017年第1期102-106,共5页 Application Research of Computers
关键词 复制 再生码 修复入度 修复带宽 磁盘I/O replication regenerating codes repair degree repair bandwidth disk I/O
  • 相关文献

参考文献2

二级参考文献81

  • 1Layman P, Varian H R. How much information 2003? [EB/OL]. [2010 10-18]. http://www2, sims. berkeley. edu/research/proiects/how-mueh-info-2003.
  • 2Pinheiro E, Weber W D, Barroso L A. Failure trends in a large disk drive population [C] //Proc of the 5th USENIX Conf on File and Storage Technologies. Berkeley, CA: USENIX Association, 2007 : 17-28.
  • 3Schroeder B, Gibson G A. Disk failures in the real world: What does an MTTF of 1,000,000 hours mean to you? [C] //Proc of the 5th USENIX Conf on File and Storage Technologies. Berkeley, CA: USENIX Association, 2007: 1-16.
  • 4Bairavasundaram L N, Goodson G R, Pasupathy S, et al. An analysis of latent sector errors in disk drives [C]//Proc of 2007 ACM SIGMETRICS Int Conf on Measurement and Modeling of Computer Systems. New York: ACM, 200: 289-300.
  • 5Hafner J M, Deenadhayalan V, Rao K, et al. Matrix methods for lost data reconstruction in erasure codes [C] // Proc of the 4th USENIX Conf on File and Storage Technologies. Berkeley, CA: USENIX Association, 2005: 183-196.
  • 6Hafner J M, Deenadhayalan V, Kanungo T, et al. Performance metrics for erasure codes in storage systems, RJ 10321 [R]. San Jose, [A] IBM Research, 2004.
  • 7Li M, Shu J, Zheng W. GRID Codes: Strip based erasure codes with high fault tolerance for storage systems [J].ACM Transon Storage, 2009, 4(4): 1-22.
  • 8Blaum M, Brady J, Bruek J, et al. EVENODD: An efficient scheme for tolerating double disk failures in RAID architectures [J].IEEE Trans on Computer, 1995, 44 (2) 192-202.
  • 9Corbett P, English B, Goel A, et al. Row-diagonal redundant for double disk failure correction [C] //Proc of the 3rd USENIX Conf on File and Storage Technologies. Berkeley, CA: USENIX Association, 2004:2-15.
  • 10Xu L, Bruck J. X-code: MDS array codes with optimal encoding[J]. IEEE Trans on Information Theory, 1999, 45 (1) : 272-276.

共引文献101

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