Storage Controller Power Failure Data Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage apparatuses with single controllers face challenges in preventing cache data loss during power failures, especially when power recovery processing is incomplete, and the power stored in capacitors may not be sufficient for re-backup operations.
Innovation Solution
Incorporating a backup processing unit to store data in a second memory during power failures, a restore processing unit to re-establish data in the first memory, an erasure processing termination unit to halt erasure during power failures, and a re-backup processing unit to re-backup data from the first memory to the second memory starting from the last unerased location, ensuring data integrity and efficient flash memory utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If power recovery processing is performed after erasure processing using capacitor power, then cache data can be restored, but the capacitor power may be insufficient for re-backup operations after erasure
Solution Approach 1:
The system performs power failure processing (backing up cache data to flash memory) before power recovery processing. This preliminary backup ensures that even if power is lost during erasure or re-backup operations, the cache data is already preserved in flash memory, preventing data loss.
Solution Approach 2:
The system uses capacitor power to perform critical operations (power failure processing and power recovery processing) before the main power is restored. This cushioning approach ensures that cache data is backed up and restored using reliable capacitor power before attempting re-backup operations that may require additional power.
2Productivity
If erasure processing is completed for all flash memory blocks, then flash memory is cleaned for new backups, but power may be exhausted before re-backup can occur
Solution Approach 1:
The system performs power failure processing to backup cache data to flash memory before erasure processing. This ensures that even if power is lost during erasure or before re-backup, the cache data is already safely stored in flash memory, preventing data loss while maintaining flash memory utilization.
3Device complexity
If a single controller is used in RAID apparatus, then device complexity is reduced, but the ability to handle power failures during erasure processing is compromised
Solution Approach 1:
The single controller is designed to autonomously handle power failure scenarios by implementing power failure processing that backs up cache data to flash memory before erasure. This self-service capability allows the controller to protect against power failures without requiring redundant controllers, maintaining simplicity while ensuring reliability.
Data Source
AI summary
A storage apparatus includes a backup processing unit that stores data stored in a first memory into a second memory as backup data upon occurrence of a power failure, a restore processing unit that upon recovery from the power failure restores the backup data backed up in the second memory to the first memory and erases the backup data, and an erasure processing termination unit that terminates the erasure processing upon a power failure occurring during erasure processing for erasing the backup data stored in the second memory, and a re-backup processing unit that re-backs up data in the first memory corresponding to the backup data erased from the second memory before the erasure processing is terminated by the erasure processing termination unit to a location in the second memory subsequent to a last location that contains the backup data which has not been erased.