Distributed Storage Recovery Using Symmetric Triple-Node Reconstruction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing distributed storage systems have relatively low data recovery performance, particularly when three nodes are lost, due to inefficiencies in existing redundancy algorithms like Reed Solomon Code and EVENODD/STAR, which result in high storage costs and complex recovery processes.

Innovation Solution

A data recovery method that determines a target data storage node based on symmetry and uses optimized cross transposition operations with diagonal and inverse diagonal parity data to recover data without loss, improving recovery performance by transforming data into XOR sums across multiple nodes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If Reed Solomon Code algorithm is used to improve storage space utilization, then storage efficiency is improved, but data recovery performance deteriorates due to low multiplication performance on CPU

Engineering Contradiction:

Improvestorage space utilizationVSAvoiddata recovery performance

Core Design Contradiction:

Quantity of substanceVSProductivity

Solution Approach 1:

The patent replaces the traditional Reed Solomon Code multiplication operation with an optimized algorithm that uses addition and bitwise operations instead of multiplication. This substitution maintains the error correction capability while dramatically improving CPU processing performance during data recovery operations.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the computational parameters of the erasure code algorithm by implementing a modified calculation approach that operates on extended bit widths (beyond the traditional 32-bit limit). This allows for more efficient processing and higher performance while maintaining storage space utilization benefits.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If Reed Solomon Code is used to extend bit width beyond 32 bits, then performance is improved, but the algorithm complexity increases

Engineering Contradiction:

Improvedata recovery performanceVSAvoidalgorithm complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the large bit-width data processing into smaller manageable units that can be processed through a series of simpler operations. By breaking down the extended bit-width calculations into sequential addition and bitwise operations, the algorithm achieves high performance without requiring complex multiplication operations.

Inventive Principle:

Principle #1Segmentation

3Reliability

If EVENODD and STAR recovery algorithms are used to recover data when three data storage nodes are lost, then data recovery is possible, but the recovery process becomes complex and difficult to implement by coding

Engineering Contradiction:

Improvedata recovery capabilityVSAvoidimplementation ease

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent inverts the traditional recovery approach by first recovering data on a target storage node using simplified calculations, then using that recovered data to reconstruct the remaining lost data. This reverse approach simplifies the coding implementation compared to traditional methods that attempt to recover all lost nodes simultaneously through complex systems of equations.

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent performs preliminary recovery of data on a target storage node before completing the recovery of all lost data. This preliminary action creates a foundation that simplifies subsequent recovery operations, making the overall process easier to implement through coding compared to attempting full recovery in a single complex step.

Inventive Principle:

Principle #10Preliminary action

4Reliability

If horizontal parity node recovery is performed first when two data storage nodes and a horizontal parity node are lost, then data recovery can proceed, but recovery performance is reduced

Engineering Contradiction:

Improvedata recovery capabilityVSAvoidrecovery performance

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent inverts the traditional recovery sequence by recovering data on a target storage node first, rather than recovering the horizontal parity node first. This inversion improves recovery performance by enabling more efficient calculation operations that leverage the structure of the storage system and the distribution of parity data across nodes.

Inventive Principle:

Principle #13The other way round (Inversion)

Data Source

PatentUS9529675B2Data recovery method, data recovery device and distributed storage system

Publication Date: 2016.12.27 HUAWEI TECH CO LTD

AI summary

A data recovery method, a data recovery device and a distributed storage system are provided, where the method includes: in a case that a distributed storage system loses data of three nodes, recovering data on a target data storage node of the data on the three nodes according to data of a parity node and a data storage node without data loss; and performing degraded recovery on the remaining lost data according to the recovered data of the target data storage node. According to the embodiments of the present invention, a target data storage node first recovered is determined according to the symmetry of lost data, and the lost data of three nodes is recovered according to parity data and data that is not lost, which can improve the data recovery performance of the distributed storage system in a case that the data of three nodes is lost.