Replication Volume Consistency via Signature Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data protection systems face challenges in ensuring consistency between production and replication volumes, particularly when these volumes are dynamically changing or rolling back to different points in time, as they often require system shutdowns, are limited in recovery points, and have lengthy recovery processes.
Innovation Solution
A method is introduced to create production and replication change maps, transferring signatures between volumes while excluding pending IO, and comparing these signatures to determine consistency, allowing for dynamic verification of replicated drives even during updates or rollbacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tape backup drives are used for periodic data storage, then data protection is provided, but system shutdown is required during backup and data recovery is limited to specific time points
Solution Approach 1:
The patent implements continuous data protection through block-level replication that operates continuously without requiring system shutdown. The replication process copies data blocks from the production storage system to the backup storage system in real-time, enabling both systems to remain operational while maintaining data protection.
Solution Approach 2:
The patent replaces the mechanical tape backup system with a digital block-level replication system. Instead of using physical tape drives that require shutdown, the system uses software-based replication at the block level, allowing continuous operation and flexible recovery at any point in time.
2Reliability
If daily data backup is performed, then data protection is maintained, but several hours of lost data occur during disaster recovery
Solution Approach 1:
The patent enables continuous data protection where data is replicated continuously rather than periodically. This allows recovery at any point in time within the recent history, eliminating the several hours of data loss that occur with daily backups. The continuous replication ensures minimal data loss window.
Solution Approach 2:
The patent performs preliminary replication of data blocks before potential disasters occur. By maintaining continuous replication, the system prepares backup data in advance, enabling rapid recovery without waiting for the next scheduled backup cycle.
3Reliability
If conventional data replication is used, then continuous data protection is achieved, but consistency verification is not possible during dynamic changes
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors and verifies consistency between production and replicated volumes. The system compares signatures of data blocks and provides feedback when inconsistencies are detected, enabling verification during dynamic changes without interrupting replication.
Solution Approach 2:
The patent replaces traditional mechanical consistency checking with signature-based verification. By using cryptographic signatures or hash values of data blocks, the system can efficiently verify consistency during dynamic changes without requiring system shutdown or interrupting replication operations.
4Reliability
If journaling is used for continuous data protection, then rollback to any point in time is enabled, but system complexity increases
Solution Approach 1:
The patent extracts the journaling functionality from the broader replication system. Instead of implementing full journaling with undo information, the system uses a simplified approach where only the essential replication metadata and block signatures are maintained, reducing complexity while preserving rollback capability.
Solution Approach 2:
The patent uses copying of data block signatures rather than full journaling. By maintaining copies of block signatures and using these for verification and rollback, the system achieves point-in-time recovery without the complexity of traditional journaling mechanisms.
Data Source
AI summary
A method, system, and program product for ensuring consistency of dynamically changing volumes comprising creating a production change map for a first volume; wherein the production change map is marked to represent pending IO, creating a replication change map for a second volume; wherein the replication change map is marked to represent pending IO, transferring a first signature corresponding to a portion of the first volume to the replication volume, wherein the signature does not include the portion marked as pending IO in the production change map, comparing the first signature to a second signature, the second signature corresponding to a portion of the replication volume corresponding to the portion of the production volume; wherein portions marked as dirty in the replication change map are not included in the signature.