Flexible Scanning for Asynchronous Replication Volume Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data replication methods face challenges in altering consistency groups undergoing asynchronous replication, particularly in adding or recovering storage volumes that have failed consistency checks during ongoing replication processes.
Innovation Solution
The implementation of flexible scanning functionality allows for efficient storage volume addition and recovery in asynchronous replication by scheduling differential and full scans within ongoing replication cycles, using snapshot trees with identified start and stop nodes, and clearing full scan flags upon completion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional data replication methods are used for ongoing asynchronous replication, then data consistency is maintained, but storage volume addition and recovery operations become difficult and require temporary sessions with multiple independent coordination steps
Solution Approach 1:
The system dynamically adjusts the scanning approach for each snapshot tree based on its specific state. Some snapshot trees undergo differential scanning while others undergo full scanning, allowing the replication process to adapt to individual volume conditions without requiring temporary sessions or complex coordination
Solution Approach 2:
The consistency group is divided into individual snapshot trees, each representing a storage volume or portion thereof. This segmentation allows independent processing of each snapshot tree with appropriate scanning depth, enabling volume addition and recovery operations without affecting the entire consistency group
2Reliability
If full scan is performed for all snapshot trees in ongoing replication, then data consistency is ensured, but replication efficiency and productivity decrease
Solution Approach 1:
Different scanning depths are applied to different snapshot trees based on their specific requirements. Snapshot trees representing volumes that have been added or recovered undergo full scanning to ensure consistency, while other snapshot trees undergo differential scanning to maintain replication efficiency
Solution Approach 2:
The scanning parameter (depth) is changed based on the snapshot tree's state. The system determines whether to perform a full scan or differential scan for each snapshot tree, dynamically adjusting the replication process to balance consistency requirements with efficiency
3Productivity
If differential scan is used for snapshot trees of added storage volumes, then replication efficiency is maintained, but data consistency may be compromised for volumes without previous scan history
Solution Approach 1:
The system performs a full scan as a preliminary action for snapshot trees representing added or recovered volumes before transitioning to differential scanning. This preliminary full scan establishes the baseline consistency needed for subsequent differential scans to be effective
Data Source
AI summary
An apparatus is configured to generate a current snapshot set for a consistency group comprising a plurality of storage volumes subject to replication from a source storage system to a target storage system, and to schedule a differential scan of the current snapshot set relative to a previous snapshot set generated for the consistency group. For each snapshot tree of a first group of one or more snapshot trees of the current snapshot set, a non-root node is identified as a start node. For each snapshot tree of a second group of one or more other snapshot trees of the current snapshot set, a root node of that snapshot tree is identified as a start node. Differential scan is performed for each snapshot tree of the first group, and full scan is performed for each snapshot tree of the second group.