通信工程文献翻译——摘要：本文考虑了一种分布式存储系统，可以重构多个存储节点同时在一个集中的位置。这种集中式多节点修复（CMR）模型是一个泛化的再生代码，允许对单个故障节点进行带宽有效的修复。这项工作的重点是在此CMR模型中存储的数据量和修复带宽之间的折衷。尤其是，为最小存储多节点修复（MSMR）和最小值修复带宽边界带宽多节点修复（MBMR）工作点。这些边界的紧密度通过分析代码结构。MSMR点的特点是通过功能实现这一点的代码修复一般的CMR参数，以及能够对某些CMR进行精确修复的代码参数。另一方面，MBMR点的特征在于所有CMR的精确修复代码满足一定熵积累性能的系统的参数。

关键词：分布式存储代码，再生码，协同多节点再生码。

一 介绍

保存存储信息并维护无缝操作的能力永久性故障和（或）存储节点的暂时不可用性是最重要的问题之一。在设计分布式存储系统时需要解决。这产生了所谓的“代码”修复“或”节点修复“问题，需要存储系统使机制重新生成（修复）存储在一些（故障/不可用）存储节点上的内容借助存储在其上的内容系统中的剩余（实时/可用）节点。一个简单的复制方案，其中存储多个副本在不同节点上的每个数据块清楚地使节点修复能够重新生成数据块通过从系统中的其他节点获得其副本之一来存储在节点上。但是，复制由于增加复制因子以增加弹性，所以具有降低的速率的系统。这导致使用擦除代码，因为它们有效地折衷了存储空间能够容忍存储节点的故障/不可用性。但是，更好的利用存储空间还应该伴随着资源节约的节点修复过程和节点修复的效率成为实现一个擦除代码超过另一个的标准。

为此，Dimakis等人提出维修带宽，从联系下载的数据量节点在修复单节点时，作为测量[1]中修复过程的效率。考虑到n个存储节点，其中任何k个节点的集合足以重建整个信息，Dimakis等人进一步表征存储空间与修复带宽之间的信息理论权衡对于这样的代码。在这种折上达到任何一点的代码被称为再生代码。在过去几年中，设计再生代码的问题促成了许多研究工作这导致了[1],[2]，[3]，[4]，[5]及其中的参考文献中提出的结构。

在本文中，我们探讨了在多个节点中实现带宽有效修复的问题集中的式。特别地，我们考虑一个设置，其中需要n中的任何k的内容，统中的节点足以重建整个信息（作为最差的参数系统的情况错）。对于集中维修过程，我们考虑一个框架t≥1节点故障的修复是......

Abstract—This paper considers a distributed storage system,where multiple storage nodes can be reconstructed simultaneously at a centralized location. This centralized multi-node repair (CMR) model is a generalization of regenerating codes that allow for bandwidth-efficient repair of a single failed node.This work focuses on the trade-off between the amount of data stored and repair bandwidth in this CMR model. In particular,repair bandwidth bounds are derived for the minimum storage multi-node repair (MSMR) and the minimum bandwidth multinode repair (MBMR) operating points. The tightness of these bounds are analyzed via code constructions. The MSMR point is characterized through codes achieving this point under functional repair for general set of CMR parameters, as well as with codes enabling exact repair for certain CMR parameters. The MBMR point, on the other hand, is characterized with exact repair codes for all CMR parameters for systems that satisfy a certain entropy

accumulation property.

Index Terms—Codes for distributed storage, regenerating codes, centralized multi-node regeneration.

I. INTRODUCTION

The ability to preserve the stored information and maintain the seamless operation in the event of permanent failures and (or) transient unavailability of the storage nodes is one of the most important issues that need to be addressed while designing distributed storage systems. This gives rise to the so called ‘code repair’ or ‘node repair’ problem which requires a storage system to enable mechanism to regenerate (repair) the content stored on some (failed/unavailable) storage nodes with the help of the content stored on the remaining (live/available) nodes in the system. A simple replication scheme where one stores multiple copies of each data block on different nodes clearly enables the node repair as one can regenerate the data blocks stored on a node by obtaining one of their copies from the other nodes in the system. However, replication suffers from the decreasing rate as one increases the replication factor in order to enhance the resilience of the system. This motives the use of erasure codes as they efficiently trade-off the storage space for the ability to tolerate failure/unavailability of storage nodes. However, the better utilization of the storage space......