首页> 外国专利> Erasure coding techniques and storage area allocation for storage system with fault tolerance and scalability

Erasure coding techniques and storage area allocation for storage system with fault tolerance and scalability

机译:具有容错能力和可扩展性的存储系统的擦除编码技术和存储区域分配

摘要

Method for configuring a storage system resiliency extensible. The method, the method comprising: forming a storage system are collectively stored number of elements D, the steps of constructing the phase D is 2 or more, a storage area allocation table containing the table-registration items, and each storage for an element, the method for calculating the available capacity that depends on the constraints on the arrangement of the redundant data in the storage system, comprising the steps of: summing the available capacity, to form a total available capacity of the storage system, assigning a registration table items of the storage area allocation table, include, identifying steps each one of the storage elements. In this method, the available capacity of each storage element, when used as part of a storage system, it is its effective storage capacity, all of the storage elements, D, of forming a storage system, have the same available capacity It is not in, further, the percentage of full table identifies the registration item is dependent on its percentage of the total available capacity of one of the storage elements. By encoding the data block, a method for despite the modification of the intentional or accidental, and transmits or stores it exactly. The method, the method comprising: constructing an initial component of a number greater than 1 n, which is also derived from the block, respectively, to which the steps smaller than the block, combining the initial component, respectively, and a step of constructing a new component of greater than 1 m. In this method, each new component is smaller than the sum of the size of the initial components are combined to produce it, block it, the n set of any selected from the new component and the initial component It is possible to reconstruct the different components, further, a set of new components as different components of n selected from the initial components, the redundant block information from the initial set of n components including.
机译:配置存储系统弹性可扩展的方法。该方法,该方法包括:形成一个存储系统,该存储系统中总共存储了元素D的数量,构造阶段D的步骤为2或更大,包含表注册项的存储区域分配表,以及每个元素的存储,取决于存储系统中冗余数据的排列约束的计算可用容量的方法,包括以下步骤:对可用容量求和,以形成存储系统的总可用容量,分配注册表项该存储区域分配表的步骤包括识别每个存储元件的步骤。在此方法中,每个存储元件的可用容量在用作存储系统的一部分时是其有效存储容量,构成存储系统的所有存储元件D都具有相同的可用容量。此外,完整表的百分比表示注册项取决于其在存储元素之一的总可用容量中的百分比。通过对数据块进行编码,该方法可以不管有意还是无意地进行修改,并精确地传输或存储它。该方法,该方法包括:构造数量也大于1 n的初始分量,该初始分量也分别从该块导出,并且小于该块的步长也分别从该块导出,并分别组合初始分量和构造步骤大于1 m的新分量。在这种方法中,每个新的分量都小于初始分量的总和,以产生,阻塞,从新分量和初始分量中选出n组任意一个。可以重构不同的分量进一步地,从初始分量中选择作为n个不同分量的一组新分量,来自n个分量初始集合的冗余块信息包括。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号