Consistency Checkpoints for Mass Storage Data Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data recovery methods in mass storage systems face challenges in ensuring data integrity and consistency during system failures, as existing techniques require synchronization of snapshots, checkpoints, and transactions, which can lead to data inconsistency and incomplete recovery.
Innovation Solution
The method generates consistency checkpoints and associates them with global snapshot numbers, using an access-based promotion technique to maintain an invariable order of checkpoints and dirty data portions, allowing for independent checkpoint management and eliminating the need for synchronization during recovery, relying on existing cache functionalities and resilience mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronization of snapshots, checkpoints, and transactions is used to ensure data consistency, then data integrity is improved, but system complexity and recovery time increase
Solution Approach 1:
The patent extracts the synchronization requirement by introducing consistency checkpoints that are automatically associated with snapshots. The checkpoint mechanism independently tracks data consistency without requiring manual synchronization coordination, thereby reducing operational complexity while maintaining data integrity.
Solution Approach 2:
The consistency checkpoint acts as an intermediary between snapshots and transactions. It provides a standardized reference point that automatically links to the corresponding snapshot, eliminating the need for direct synchronization between multiple components and simplifying the recovery process.
2Reliability
If synchronization of snapshots, checkpoints, and transactions is implemented, then data consistency is improved, but recovery time increases
Solution Approach 1:
Consistency checkpoints are created in advance during normal operation, establishing known good states before failures occur. During recovery, the system can immediately use these pre-created checkpoints as reference points, eliminating the need to wait for synchronization processes to complete and significantly reducing recovery time.
Solution Approach 2:
The patent creates a copy of the consistency state through checkpoints that are independently stored with snapshots. This copying allows the system to recover by restoring from the checkpoint copy without requiring complex synchronization operations, thereby reducing recovery time while maintaining data consistency.
3Reliability
If access-based promotion technique is used for cache management, then data recovery consistency is improved, but cache performance may be affected
Solution Approach 1:
The patent applies different handling rules to different data in the cache based on their association with snapshots and checkpoints. Data portions linked to snapshots are treated differently from regular data, with specific promotion restrictions applied only where needed to maintain recovery consistency, while allowing normal cache performance for other data.
Solution Approach 2:
The consistency checkpoint mechanism applies promotion restrictions only to the extent necessary for recovery consistency, not to all cache data. This partial application allows the system to maintain cache performance for the majority of data while ensuring recovery consistency for critical data portions associated with snapshots.
Data Source
AI summary
A mass storage system and method incorporates a cache memory or a cache management module which handles dirty data using an access-based promotion replacement process through consistency checkpoints. The consistency checkpoints are associated with a global number of snapshots generated in the storage system. The consistency checkpoints are organized within the sequence of dirty data in an invariable order corresponding to storage volumes with the generated snapshots, such that, responsive to destaging a consistency checkpoint the global number of generated snapshots are recorded and then read during recovery of the failed storage system.