Non-Volatile Memory Controller Safety Moving for All-Zero Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory, particularly flash memory, deteriorates due to all-0's or all-1's data, which can damage storage capability and lead to repeated duplication of undesirable data, wasting storage space.
Innovation Solution
A data storage device with a controller that performs safety moving of valid data from high-risk blocks containing all-1's or all-0's data, scrambling and descrambling data before writing and reading, and using a deterioration table to identify and manage deteriorated physical addresses, preventing duplication of undesirable data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If safety moving is performed on high-risk blocks containing all-1's or all-0's data, then storage space is protected from deterioration, but the complexity of data management increases due to the need to identify and manage deteriorated blocks
Solution Approach 1:
The controller performs preliminary identification of high-risk blocks containing all-1's or all-0's data before deterioration occurs. By detecting these patterns early and marking them as high-risk, the system proactively prevents deterioration rather than reacting after damage occurs, thus protecting storage space while managing complexity through early intervention
Solution Approach 2:
The controller acts as an intermediary between the non-volatile memory blocks and the host system. It intercepts read requests for high-risk blocks, performs safety moving operations, and returns corrected data to the host, thereby shielding the host from the complexity of deterioration management while protecting storage space
2Ease of operation
If all-1's or all-0's data is repeatedly duplicated in non-volatile memory, then data retrieval is simplified, but storage space is wasted due to deterioration
Solution Approach 1:
The controller converts the harmful effect of all-1's or all-0's data patterns into a beneficial detection mechanism. By using these distinctive patterns as markers for high-risk blocks, the system easily identifies deteriorated areas while preventing further waste, thus turning potential storage loss into an early warning system
Solution Approach 2:
When deterioration is detected in high-risk blocks, the controller discards the corrupted data by performing safety moving operations that exclude deteriorated blocks from future use. The valid data is recovered and relocated to healthy blocks, preventing further storage space waste while maintaining data accessibility
3Measurement precision
If the controller performs error checking and correction on read data, then data accuracy is improved, but the processing time increases
Solution Approach 1:
The controller performs preliminary error checking and correction operations on high-risk blocks before they are fully processed or returned to the host. By pre-processing these problematic blocks, the system ensures data accuracy is established early, preventing later re-processing and reducing overall processing time
Solution Approach 2:
For high-risk blocks that fail error checking, the controller quickly skips normal processing routines and immediately initiates safety moving operations. This expedited handling of deteriorated data prevents processing delays from propagating through the entire data retrieval pipeline, maintaining overall system efficiency
Data Source
AI summary
A solution for deteriorated non-volatile memory is shown. When a controller determines that raw data read from the non-volatile memory is undesirable data, the controller performs safety moving of valid data of an erasure unit that contains the raw data to safely move the valid data of the erasure unit, wherein the erasure unit is a high-risk block, and the raw data in the non-volatile memory is regarded as being in a deteriorated physical address. Prior to being moved in the safety moving, the raw data is changed so that it is different from the undesirable data. In an exemplary embodiment, the undesirable data is all-1's data or all-0's data.