Logical-to-Physical Table Recovery Using Candidate Codeword Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory systems face challenges in recovering logical-to-physical table information when errors occur in error correcting and detecting codes, leading to difficulties in identifying physical locations of data and resulting in system failures or long latencies.
Innovation Solution
The implementation of a method that determines candidate codewords with a Hamming distance equal to the bit errors in corrupted codewords, leveraging metadata and the sequential nature of L2P pointers to quickly identify correct candidate codewords, thereby reducing latency in recovering corrupted L2P information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard error correcting codes are used to protect L2P table information, then reliability is improved, but the system cannot recover from uncorrectable errors and requires full storage space scanning which increases latency
Solution Approach 1:
The patent pre-generates candidate codewords based on the detected error pattern (Hamming distance) before actual recovery is needed. When an uncorrectable error occurs, the system already has a sorted list of candidate codewords to evaluate, eliminating the need to scan the entire storage space and significantly reducing recovery latency
Solution Approach 2:
The patent segments the recovery process into discrete candidate codewords based on Hamming distance from the corrupted codeword. Instead of treating recovery as a single exhaustive search, it divides the solution space into manageable candidates that can be evaluated in sequence, with the most promising candidates checked first
2Reliability
If the entire storage space is scanned to recover corrupted L2P information, then reliability is improved by ensuring thorough search, but productivity deteriorates due to long scanning time
Solution Approach 1:
The system performs preliminary generation of candidate codewords using the known error pattern (Hamming distance) before the actual recovery operation. This pre-computation creates a sorted list of candidates that can be quickly evaluated, maintaining recovery completeness while dramatically improving speed by avoiding full storage space scanning
Solution Approach 2:
The patent changes the approach from scanning physical storage locations to evaluating codeword parameters in the logical address space. By working with candidate codewords and their Hamming distances rather than physically scanning storage cells, the system achieves the same recovery goal with much higher efficiency
3Reliability
If complex error correction procedures are implemented to handle uncorrectable errors, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent uses feedback from the error detection stage (Hamming distance calculation) to guide the recovery process. The detected error pattern directly informs which candidate codewords to generate and in what order to evaluate them, creating an adaptive recovery mechanism that responds to the specific error condition rather than using a fixed complex procedure
Data Source
AI summary
Methods, systems, and devices for recovery of logical-to-physical (L2P) table information for a memory device are described. A memory system may detect an error in one or more pointers of the L2P table using an error detecting code that is uncorrectable using the code. The memory system may determine a set of candidate codewords for the set of bits, where each of the candidate codewords includes one or more corresponding candidate pointers, and check whether a candidate codeword is correct based on whether a logical address corresponding to a candidate pointer of the candidate codeword matches a logical address stored as metadata for a set of data at a physical address pointed to by the candidate pointer. The memory system may limit the set of candidate codewords or order the candidate codewords for evaluate to reduce a latency associated with identifying a correct candidate codeword.