Mirrored Volume Resynchronization via Post-Failure Snapshot
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resynchronizing mirrored volumes across different sites in storage systems can be slow due to bandwidth constraints, leading to inconsistent data and potential transactional inconsistencies, especially during failures and recovery processes.
Innovation Solution
A system and method where a storage management device at a primary site conveys an indication of data blocks written subsequent to a failure to a secondary site, allowing the creation of a snapshot before resynchronization, thereby reducing the need for repeated data transfers over a potentially bandwidth-limited link and ensuring data consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resynchronization is performed by copying all data blocks from primary volume to secondary volume, then data consistency is restored, but resynchronization time increases due to bandwidth constraints
Solution Approach 1:
The system performs preliminary actions by identifying and marking data blocks that were written to the primary volume after the link failure occurred. This metadata about which blocks need resynchronization is prepared and communicated to the secondary site before the actual resynchronization copying begins, enabling more efficient targeted replication rather than copying all data.
Solution Approach 2:
The invention extracts only the necessary data blocks that require resynchronization from the primary volume, rather than transferring the entire volume. By using the indication of post-failure writes, the system identifies and extracts only the affected blocks for replication, significantly reducing the data transfer volume and time required.
2Reliability
If all data blocks are transferred during resynchronization, then transactional consistency is ensured, but communication resources are inefficiently utilized
Solution Approach 1:
The system applies partial action by transferring only the subset of data blocks that actually changed after the failure, rather than transferring all blocks. The indication mechanism provides metadata about which blocks are affected, allowing the secondary site to perform partial resynchronization only on necessary blocks, optimizing communication resource usage.
Solution Approach 2:
The primary storage management device provides feedback in the form of an indication about which data blocks were written after the link failure. This feedback information is conveyed to the secondary site to enable intelligent resynchronization decisions, ensuring that only the necessary blocks are transferred and processed, thereby improving communication efficiency.
Data Source
AI summary
System and method for resynchronizing mirrored volumes in a storage system. According to one embodiment, a system may include a first volume located at a primary site and associated with a first storage management device, and a second volume located at a secondary site and associated with a second storage management device, where the secondary site is coupled to the primary site via a data link, and where the second volume is configured as a mirror of the first volume. The first storage management device may be configured to convey to the second storage management device an indication of a data block written to the first volume subsequent to a failure of the data link. Dependent upon the indication, the second storage management device may be configured to create a snapshot of the data block as stored on the second volume prior to resynchronization of the data block.