Flash Memory Controller Sudden Power-Off Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data storage devices face performance degradation and increased erase counts due to sudden power-offs during data transfer, leading to discarded data and reduced operation efficiency.
Innovation Solution
A data processing method that identifies and recovers affected pages by double programming directly related pages and dummy programming indirectly related pages, allowing for resumed data transfer without discarding successfully transferred data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data storage devices continue normal operation after sudden power-off, then device availability is maintained, but data integrity deteriorates due to corrupted pages from incomplete write operations
Solution Approach 1:
The system performs preliminary identification of affected pages by checking status flags before resuming operations. This preliminary action allows the device to detect which pages were corrupted during sudden power-off events and prepare appropriate recovery actions, thereby maintaining data integrity while enabling continued operation.
Solution Approach 2:
The system identifies corrupted pages through status flags and applies targeted recovery operations (double programming or dummy programming) only to affected pages rather than discarding entire blocks. This selective recovery approach preserves data integrity while minimizing impact on overall device availability and productivity.
2Reliability
If all pages are re-programmed after sudden power-off, then data integrity is ensured, but operation time increases due to unnecessary re-programming of unaffected pages
Solution Approach 1:
The system segments the recovery process by identifying and treating only the specific pages affected by sudden power-off events.通过使用状态标志位，系统能够将 affected pages 与 unaffected pages 分离开来，仅对前者进行双次编程或虚拟编程操作，从而在保证数据完整性的同时显著减少操作时间.
Solution Approach 2:
The system applies different quality levels of recovery operations to different pages based on their actual state. Affected pages receive intensive recovery treatment (double programming), while unaffected pages receive minimal or no treatment. This localized approach ensures data integrity where needed while avoiding unnecessary time consumption on already healthy pages.
3Productivity
If erase operations are performed frequently to maintain performance, then operation performance is improved, but device lifespan deteriorates due to increased erase counts
Solution Approach 1:
Instead of performing full erase operations on entire blocks when pages are affected, the system recovers only the specific corrupted pages through targeted double programming or dummy programming. This page-level recovery approach maintains operation performance by keeping affected pages usable while avoiding the performance-degrading full block erasures that would reduce device lifespan.
Solution Approach 2:
The system converts the harmful effect of sudden power-off (which corrupts pages) into a beneficial opportunity for targeted recovery. By using the status flags left by power-off events to identify exactly which pages need recovery, the system performs minimal necessary operations rather than extensive erasures, thus maintaining performance while extending device lifespan.
4Manufacturing precision
If status flags are used to identify affected pages, then manufacturing precision is improved through accurate page identification, but device complexity increases due to additional flag management mechanisms
Solution Approach 1:
The system uses self-service mechanisms where the sudden power-off event itself leaves status flags that automatically identify affected pages. This self-identifying approach achieves high page identification accuracy without requiring complex external detection systems, as the pages essentially tag themselves through the flag status during the power-off event.
Solution Approach 2:
The system merges the status flag management with the existing page structure and control logic. Rather than adding completely separate flag management subsystems, the flags are integrated into the existing memory control architecture, allowing accurate page identification while minimizing additional device complexity through unified management.
Data Source
AI summary
A data storage device includes a memory device and a controller. The memory device includes at least an MLC block. The MLC block includes a plurality of physical pages. The controller is coupled to the memory device. When the controller determines that a sudden power-off has occurred during a previous write operation for writing data onto the MLC block, the controller finds a predetermined page that has been attacked by the sudden power-off, double programs the predetermined page and a first page that is directly related to the predetermined page and dummy programs a plurality of second pages that are indirectly related to the predetermined page.