Distributed Data Storage Slice Repair for Active-Active Centers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Active-active data centers face instability when data cannot be written into either data center, leading to a loss of the active-active feature, which is critical for ensuring continuous service and disaster recovery in cloud computing and big data environments.
Innovation Solution
A data storage method that divides data into N data slices and M check slices, distributing them across R storage nodes in two data centers, allowing for fault tolerance and repair of failed slices, thereby maintaining data center reliability and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written into storage nodes in a single data center using active-active configuration, then data redundancy is improved, but system stability deteriorates when data cannot be written into either data center
Solution Approach 1:
The invention segments data into N data slices and M check slices, distributing them across R storage nodes in different data centers. This segmentation allows the system to maintain functionality even when some segments are lost or corrupted, resolving the contradiction between achieving redundancy and maintaining stability.
Solution Approach 2:
The invention performs preliminary actions by pre-distributing data slices and check slices across multiple data centers before failures occur. The check slices are prepared in advance to enable repair operations, ensuring the system can recover from failures without losing the active-active feature.
2Reliability
If data is divided into multiple slices and distributed across storage nodes, then fault tolerance is improved, but system complexity increases
Solution Approach 1:
The invention creates M check slices as copies derived from N data slices through encoding operations. These check slices serve as redundant copies that can be used to repair lost data slices, achieving fault tolerance while managing complexity through systematic copying rather than full duplication.
Solution Approach 2:
The invention changes the parameter representation of data by transforming original data into N data slices and M check slices through mathematical encoding. This parameter transformation enables fault tolerance while maintaining manageable system complexity through structured mathematical relationships.
3Ease of repair
If check slices are generated and stored alongside data slices, then data repair capability is improved, but storage space requirements increase
Solution Approach 1:
The invention implements a strategy where M check slices are stored alongside N data slices. When data slices are lost or corrupted, the system recovers the original data by using the check slices in combination with remaining data slices, effectively discarding the need to store every single original copy while maintaining repair capability.
Data Source
AI summary
A data storage method and an apparatus are provided in a distributed storage system including a computing node and a plurality of storage nodes. The computing node writes the N data slices and the M check slices into the R storage nodes in each storage node group to improve reliability and stability of data in a data center.