Storage Testing Method Using Segmented Error Bit Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for testing storage apparatuses, such as memory devices, require writing and reading large amounts of patterns, consuming time and reducing memory endurance due to limited access times.
Innovation Solution
A method involving writing a specific pattern to a storage unit, reading it, and determining defects based on error bit numbers compared to a threshold value, which is set below the correctable bit number for error correction codes, allowing for efficient testing with fewer patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large amount of patterns are written to storage device for testing, then testing coverage and reliability are improved, but testing time increases and memory endurance is reduced
Solution Approach 1:
The testing process is segmented into two distinct stages: (1) a rough testing stage using a first specific pattern to quickly identify obviously defective storage units, and (2) a precise testing stage using a second specific pattern to accurately determine defects in non-obviously defective units. This segmentation allows the system to achieve comprehensive testing coverage while minimizing the total number of read-write operations required.
Solution Approach 2:
The invention applies partial action by using a threshold value for error bit numbers to determine when precise testing is necessary. Not all storage units require the same level of testing intensity - obviously defective units are identified quickly with the first pattern, while only non-obviously defective units undergo the more intensive second pattern testing. This partial application of full testing to only necessary cases reduces overall testing time and memory stress.
2Reliability
If large amount of patterns are written to storage device for testing, then testing coverage and reliability are improved, but memory endurance is reduced due to limited access times
Solution Approach 1:
The testing methodology segments storage units into different categories based on error characteristics: obviously defective storage units (with error bits exceeding the threshold) and non-obviously defective storage units (with error bits at or below the threshold). This segmentation enables tailored testing approaches that minimize read-write operations on each category, thereby preserving memory endurance while maintaining comprehensive testing coverage.
Solution Approach 2:
The invention changes the parameter of pattern characteristics between two testing stages. The first specific pattern is designed to reveal obvious defects, while the second specific pattern is optimized for detecting subtle defects. By changing pattern parameters rather than repeatedly using the same exhaustive patterns, the system achieves thorough testing coverage with fewer access operations, thus extending memory endurance.
3Measurement precision
If error bit threshold value is set below correctable bit number, then testing precision is improved, but number of patterns needed increases
Solution Approach 1:
The threshold value approach implements partial action by setting a criterion (error bit threshold) that is lower than the maximum correctable bit number. This allows the system to quickly identify and flag obviously defective storage units with high precision using the first specific pattern, while only requiring the second pattern for units that fall into the gray area. This partial application of strict threshold testing minimizes the total number of patterns needed while maintaining high detection precision.
Data Source
AI summary
A method for testing a storage apparatus, which includes: (a) writing a specific pattern to a storage unit of a storage apparatus; (b) reading the specific pattern written to the storage apparatus; (c) determining an error bit number of the specific pattern read in the step (b); and (d) determining that the storage unit has defect when the error bit number is larger than a error bit threshold value, wherein the error bit threshold value is smaller than a correctable bit number for a error correction code corresponding to the specific pattern.