Storage Cache Track Swapping for Mistaken Point-in-Time Copy Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for recovering from unintentional or mistaken point-in-time copy restores in enterprise-level storage systems are time-consuming and may result in data loss, as they require an additional intermediate volume and are prone to human error, leading to extended system unavailability.
Innovation Solution
The method involves swapping tracks between a source and target volume using a storage subsystem cache, based on a change indicator, allowing data to be preserved and reversed if an incorrect volume is restored, eliminating the need for an intermediate volume and reducing human error risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If an additional intermediate volume is used to recover from a mistaken point-in-time copy restore, then data can be preserved, but device complexity and time consumption increase
Solution Approach 1:
The patent uses a change indicator bitmap as an intermediary mechanism to track which tracks have been modified since the point-in-time copy was created. This bitmap serves as a mediator between the source volume and target volume, enabling the system to identify and preserve changed tracks without requiring an additional intermediate volume. The bitmap acts as a metadata structure that facilitates the recovery process by providing information about data changes.
Solution Approach 2:
The patent extracts the change tracking functionality from the traditional intermediate volume approach and implements it through a change indicator bitmap stored in the storage subsystem cache. By extracting this specific function (tracking changes) and implementing it separately through metadata rather than through a full intermediate volume, the system achieves data preservation without the complexity of additional storage volumes.
2Reliability
If manual recovery procedures are used for mistaken restores, then data can be recovered, but loss of time and productivity increase
Solution Approach 1:
The patent implements preliminary action by automatically tracking all changes to tracks since the point-in-time copy was created using a change indicator bitmap. This preliminary tracking of modifications enables the system to quickly identify which data needs to be preserved during recovery, eliminating the need for time-consuming manual analysis and procedures. The change tracking occurs continuously in the background before recovery is needed.
Solution Approach 2:
The system performs self-service recovery by automatically using the change indicator bitmap to identify and preserve changed tracks during the restore process. The storage subsystem cache autonomously manages the recovery operation by referencing the bitmap to determine which tracks contain modifications that should not be overwritten, eliminating the need for manual human intervention and reducing recovery time.
3Productivity
If the storage subsystem cache is used for track swapping, then productivity is improved, but use of energy increases
Solution Approach 1:
The patent applies local quality by using the storage subsystem cache specifically for holding the change indicator bitmap and facilitating track swapping operations. The cache provides high-speed access to the bitmap metadata and enables rapid track exchange between source and target volumes. By localizing the cache usage to this specific function rather than using it for general-purpose storage, the system achieves high productivity for recovery operations while minimizing overall energy consumption.
Data Source
AI summary
A method, system, and program product are provided for undoing a point-in-time copy restore. Provided are swapping corresponding tracks between a source volume and a target volume using cache of a storage subsystem to which the source and the target volumes are attached, based on an indicator being set that a track on the source volume is changed since a point-in-time copy. Tracks are moved from the source volume to the target volume based on the indicator being unset.