Encoder Bypass Scrambler for Faster NAND Flash Data Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data migration techniques in solid state storage systems, such as NAND Flash, are limited in efficiency and power consumption when they incorporate scrambling mechanisms that use the same random bit sequences, and do not effectively address read disturb noise and voltage leakage issues, which can lead to incorrect bit value readings.
Innovation Solution
A data migration process that includes error correction decoding and encoding using pseudo random sequences dependent on physical locations, allowing for efficient migration by combining and summing error-corrected data with new random sequences, thereby canceling out noise and optimizing processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data migration is performed in solid state storage systems with scramblers using different random bit sequences, then data integrity can be maintained, but processing time and power consumption increase
Solution Approach 1:
The system performs preliminary scrambling of data before storage using location-dependent random bit sequences. During migration, the new data is scrambled with the same sequence before being XORed with the old data, ensuring that the scrambling operations cancel out and data integrity is maintained without requiring complex post-processing
Solution Approach 2:
The patent extracts and utilizes the linear property of XOR operations to separate the scrambling function from the migration process. By taking out the scrambling operation and applying it independently to both old and new data before combining them, the system maintains data integrity while simplifying the overall migration complexity
2Reliability
If data migration is performed in solid state storage systems with scramblers using different random bit sequences, then data integrity can be maintained, but power consumption increases
Solution Approach 1:
The system performs preliminary scrambling of data before storage using location-dependent random bit sequences. During migration, the new data is scrambled with the same sequence before being XORed with the old data, ensuring that the scrambling operations cancel out and data integrity is maintained without requiring complex post-processing
Solution Approach 2:
The patent creates a copy of the scrambling sequence based on the storage location and applies it to both the old and new data. This copying approach ensures that the same transformation is applied to both datasets, allowing them to be combined through simple XOR operations while maintaining data integrity
3Measurement precision
If read disturb noise and voltage leakage are addressed through frequent data migration, then data accuracy is maintained, but processing time and power consumption increase
Solution Approach 1:
The system uses the existing XOR operation and scrambling mechanism already present in the storage system to perform migration. The old and new data automatically cancel out the scrambling effects through XOR, making the migration process self-sufficient and eliminating the need for additional complex processing steps
Solution Approach 2:
The patent changes the parameter of the scrambling sequence to be location-dependent rather than uniform. This allows the system to maintain data accuracy through migration while optimizing processing efficiency by using the same cryptographic mechanism with varied input parameters
Data Source
AI summary
A first physical location is read to obtain read data. Error correction decoding is performed on the read data to obtain error-corrected data where the error-corrected data includes first error-corrected metadata. Error correction encoding is performed on a first random sequence combined with a second random sequence, concatenated with second metadata. Error correction encoding is also performed on a sequence of zeros concatenated with the first error-corrected metadata to obtain second encoded data. The error-corrected data, the first encoded data, and the second encoded data are summed to obtain migrated data, which is stored at a second physical location.