Read Signature Detection for ECC Mis-correction in NAND Flash
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing error correction schemes in multi-threshold level NAND flash memory fail to detect certain types of errors, leading to silent data corruption due to mis-correction during read operations, especially when read reference voltages are shifted significantly.
Innovation Solution
Implementing a read signature detection mechanism to identify ECC mis-correction by counting bits in each bucket and checking resulting signatures, which allows for the detection of extreme shifts in read reference voltages that result in ECC mis-correction and silent data corruption, while maintaining support for moving read reference voltage adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error correction coding schemes are applied to improve memory reliability, then reliability is improved, but certain types of errors still go undetected leading to silent data corruption
Solution Approach 1:
The patent segments the read verification process into multiple independent checks: traditional ECC validation and a separate read signature validation. By dividing the error detection function into distinct segments (ECC code checking and read signature pattern matching), the system can detect different types of errors that each segment is specifically designed to catch, thereby reducing silent data corruption while maintaining reliability improvements from ECC
Solution Approach 2:
The patent introduces read signatures as an intermediary mechanism between the read operation and the final data validation. These signatures act as a mediator that provides an additional layer of verification independent of the ECC scheme, allowing the system to detect mis-corrections that the primary ECC mechanism misses without compromising the reliability benefits of error correction coding
2Adaptability or versatility
If read reference voltage levels are shifted significantly to accommodate moving read reference voltage adjustments, then adaptability is improved, but ECC mis-correction occurs leading to silent data corruption
Solution Approach 1:
The patent applies preliminary action by performing read signature validation before final data acceptance. The system proactively checks for mis-correction patterns in the read signature before the erroneous data is processed or stored, allowing early detection and rejection of potentially corrupted reads resulting from significant voltage shifts, thus preventing silent data corruption while maintaining voltage adaptability
3Measurement precision
If bit counting and signature checking are implemented to detect ECC mis-correction, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent uses copying by creating and storing reference read signatures that represent expected valid patterns. Instead of implementing complex real-time analysis mechanisms, the system copies known-good signature patterns and compares incoming read signatures against these stored references, achieving high measurement precision for error detection while keeping the device complexity manageable through simple pattern matching
Data Source
AI summary
Error correction coding (ECC) mis-corrected reads, if undetected, result in silent data corruption of a non-volatile memory device. Overcoming ECC mis-corrected reads is based on a read signature of a result of reading a page in the non-volatile memory device. An ECC mis-correct logic counts the number of bits in the end-most buckets into which the bits of the result is divided. End-most buckets that are overpopulated or starved reveal a tell-tale read signature of an ECC mis-correct. The ECC mis-correct is likely to occur when the read reference voltage level used to read the page is shifted in one direction or another to an extreme amount that risks reading data from a different page. Detecting ECC mis-corrected reads can be used to overcome the ECC mis-corrects and mitigate silent data corruption.