Non-Volatile Memory Write Operation with Immediate Stale Data Destruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In non-volatile memory devices like solid state drives (SSDs), multiple versions of data sets persist over time, leading to resource-intensive sanitization processes and security risks due to the presence of stale data copies, which can be recovered by unauthorized parties.
Innovation Solution
A method where a newer copy of user data is written to a selected empty physical location while concurrently overwriting the older copy stored in a different occupied location, ensuring only one copy of a data set exists, enhancing data security and sanitization efficiency by updating the mapping table and using rubout data like all zeros to erase obsolete data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple versions of data sets are allowed to persist in non-volatile memory, then data recovery capability is improved, but data security deteriorates due to stale data copies being recoverable by unauthorized parties
Solution Approach 1:
The patent extracts and removes stale data copies from the memory system by overwriting them with rubout data during normal write operations. The mapping table is used to identify the physical locations of stale copies, which are then selectively overwritten to eliminate security risks while maintaining the ability to recover current data.
Solution Approach 2:
The patent performs preliminary overwriting of stale data copies with rubout data before potential unauthorized access can occur. By proactively destroying old data versions during write operations, the system prevents future security breaches without affecting current data accessibility.
2Object-affected harmful factors
If extensive sanitization processes are performed to remove stale data copies, then data security is improved, but resource consumption and processing time increase
Solution Approach 1:
The patent performs sanitization preliminarily by overwriting stale data copies at the time of each write operation, rather than performing batch sanitization later. This distributes the sanitization workload over time and eliminates the need for separate, resource-intensive sanitization processes.
Solution Approach 2:
The system performs self-sanitization by automatically identifying and overwriting its own stale data copies using the mapping table during normal write operations. This eliminates the need for external or scheduled sanitization processes, reducing both time and resource consumption.
3Object-affected harmful factors
If frequent erasure operations are performed to remove old data copies, then data security is improved, but wear on the non-volatile memory increases
Solution Approach 1:
The patent uses rubout data (inexpensive overwrite operation) to destroy stale data copies instead of performing costly erasure operations. The mapping table enables precise targeting of only the necessary stale copies, minimizing the number of write operations required and reducing overall memory wear.
Solution Approach 2:
The patent applies rubout data only to the specific physical locations identified in the mapping table where stale copies exist, rather than performing blanket erasure operations. This localized approach minimizes the total number of write operations and reduces wear on the non-volatile memory.
4Productivity
If mapping table is updated to track physical locations of data copies, then data management efficiency is improved, but device complexity increases
Solution Approach 1:
The mapping table serves multiple functions: it tracks current data locations for read/write operations, identifies stale copies for sanitization, and enables efficient data recovery. This multi-functionality justifies the added complexity by eliminating the need for separate tracking structures.
Solution Approach 2:
The mapping table automatically updates itself during write operations to reflect new data locations and identify stale copies. This self-maintaining mechanism reduces the need for external management overhead and simplifies the overall system architecture despite the added data structure.
Data Source
AI summary
Method and apparatus for writing data to a non-volatile memory device, such as a solid state drive (SSD). In accordance with various embodiments, a host write command is serviced by writing a newer copy of user data to a first selected empty physical location in a non-volatile memory, and by concurrently overwriting an older copy of said user data previously stored to a different, second selected occupied physical location of the non-volatile memory.