Storage Data Replication via Snapshot Metadata Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data replication methods face challenges in maintaining high availability and performance, particularly over large geographical distances, due to latency issues in synchronous replication and potential data inconsistency in asynchronous replication, as well as resource consumption and performance degradation in point-in-time replication.
Innovation Solution
A method and system for maintaining a remote replica of a storage device by storing updates in a snapshot and metadata, periodically updating the replica using the metadata, and optionally deleting the snapshot and metadata to improve efficiency and reduce resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous storage replication is used to maintain high availability, then data availability is improved, but system performance deteriorates due to latency in acknowledging write operations
Solution Approach 1:
The patent segments the replication process into two distinct phases: synchronous phase for critical data protection and asynchronous phase for continuous updates. The system maintains synchronous replication for essential data to ensure availability, while using asynchronous replication for additional data streams to avoid performance penalties, thus resolving the contradiction between reliability and productivity.
Solution Approach 2:
The patent applies partial synchronous replication by selectively synchronizing only critical data streams rather than all data. This partial action maintains high availability for essential operations while reducing the latency impact on overall system performance by excluding non-critical data from synchronous replication.
2Productivity
If asynchronous storage replication is used to improve performance over large distances, then system performance is improved, but data consistency deteriorates due to potential data loss
Solution Approach 1:
The patent segments data replication into synchronous and asynchronous channels, applying synchronous replication to critical data for consistency and asynchronous replication to additional data for performance. This segmentation allows the system to achieve both high performance and data consistency where needed.
Solution Approach 2:
The system performs preliminary synchronous replication of critical data before switching to asynchronous replication for additional data streams. This preliminary action ensures data consistency is established for essential operations before the performance-optimized asynchronous replication takes over for non-critical data.
3Quantity of substance
If point-in-time replication is used to reduce resource consumption, then resource usage is improved, but performance deteriorates due to snapshot growth
Solution Approach 1:
The patent implements dynamic snapshot management where snapshots are automatically deleted after a predetermined time period or when no longer needed. This dynamic approach allows the system to reduce resource consumption by removing old snapshots while maintaining performance by ensuring recent snapshots remain available for replication operations.
Solution Approach 2:
The system performs periodic snapshot creation followed by periodic deletion of snapshots exceeding a time threshold. This periodic action pattern allows the system to maintain necessary replication snapshots while removing obsolete ones to prevent resource exhaustion and maintain performance.
Data Source
AI summary
A method and system of replicating data stored on a storage device where an update is stored in a snapshot. The update includes a copy of at least one portion of the data stored on the storage device. Metadata relating to the update is also stored. The replica is periodically updated by copying the contents of the snapshot into the replica in accordance with the stored metadata. After the replica is updated, the snapshot can be deleted.