Unlock instant, AI-driven research and patent intelligence for your innovation.

Local Reconfiguration Code Encoding and Node Fault Repair Method in Distributed Storage System

A distributed storage and partial reconstruction technology, applied in the transmission system, digital transmission system, data exchange network, etc., can solve the problem of large repair bandwidth overhead, achieve the goals of reducing computational complexity, ensuring availability, and reducing repair bandwidth overhead Effect

Active Publication Date: 2020-09-11
CHANGAN UNIV
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The node failure repair algorithms mentioned above can only perform well for a single data node failure repair. When the check node fails, these algorithms or codes require a large repair bandwidth overhead when repairing, which leads to check The repair bandwidth overhead of a block is much greater than the repair bandwidth overhead of a data block

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Local Reconfiguration Code Encoding and Node Fault Repair Method in Distributed Storage System
  • Local Reconfiguration Code Encoding and Node Fault Repair Method in Distributed Storage System
  • Local Reconfiguration Code Encoding and Node Fault Repair Method in Distributed Storage System

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0057] like figure 2 As shown, in this embodiment, there are 21 data nodes in the distributed storage system, 12 of which are used to store 12 data blocks divided into data files, and the remaining data nodes store the first local check digit q 0 and q 1 , the second partial check p 1 to p 6 bit and the global parity bit c. Considering the impact of the recovery efficiency of a single node failure on the performance of a distributed storage system, the following mainly discusses the recovery bandwidth overhead when a single node fails:

[0058] (1) When a normal node fails

[0059] like image 3 As shown, for example, when the ordinary node d 1 If a failure occurs, the corresponding first local check node q 0 and d 1 Other ordinary nodes in the local group d 2 、d 3 、d 4 、d 5 、d 6 Repair by XOR operation; also through its corresponding second local check node p 1 and d 1 Corresponding ordinary nodes d in other local groups 7 Carry out XOR operation repair, this...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a quasi local reconstruction code encoding and node fault repairing method in a distributed storage system. The repairing features and the requirements of the distributed storage system for the node fault repairing performance are analyzed to provide the quasi local reconstruction code encoding and node fault repairing method without increasing partial storage overhead. A quasi local reconstruction code contains a global parity block generated by a local check module to ensure the repairing performance of the data node, and the data blocks of other check blocks are entirely generated by XOR operation encoding. Based on the encoding structure, only a few nodes need to be connected to achieve fast recovery in a single node fault, which ensures fast and efficient recovery of data files and reduces the repair bandwidth overhead.

Description

technical field [0001] The invention relates to the technical field of computer distributed storage, in particular to a similar local reconstruction code (SimilarLocal Reconstruction Code, SLRC) and a method for repairing node faults by using the similar local reconstruction code. Background technique [0002] In order to avoid the reduction of data transmission caused by the failure of calibration data, only by accelerating the update of calibration data, and then ensuring the stability of the system, many research works have been proposed. Among them, James S. Plank and others optimized the calculation and update of erasure code verification data based on calculation from two aspects of constructing low-density coding matrix and operation scheduling. [0003] There is also a lot of research work on faulty node repair. Xiang et al. proposed a single disk fault recovery algorithm (hybrid recovery scheme) for the RAID-6 storage system using RDP encoding. The amount of data ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): H04L1/00H04L12/24H04L29/08
CPCH04L1/0061H04L41/0663H04L67/1097
Inventor 王静张崇王珍珍刘艳余春雷
Owner CHANGAN UNIV