Flash Memory Read Disturb Mitigation via Singular Block Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash memory devices, particularly NAND flash devices, suffer from read disturbance where frequent reads of one block can cause adjacent blocks to fail due to threshold limits, necessitating methods to migrate data and prevent data loss.
Innovation Solution
A method and apparatus that utilize read-count tables to identify singularly stressed blocks, perform test reads, and migrate data from these blocks to available blocks to mitigate read disturbance, involving a processing unit, volatile memory, and access interface to manage data movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If frequent reads are performed on one physical block, then data access speed is improved, but adjacent physical blocks become programmed and fail due to read disturbance
Solution Approach 1:
The system performs preliminary actions by maintaining read-count tables that track and predict future read disturbances before they cause actual block failures. By proactively identifying blocks that will become singular based on historical read patterns, the system migrates data in advance, preventing read disturbance failures before they occur.
Solution Approach 2:
The system implements feedback mechanisms through read-count tables that continuously monitor read operations on physical blocks. This feedback information is used to predict which blocks will become singular and trigger data migration operations, creating a closed-loop system that adapts to actual usage patterns and prevents read disturbance.
2Reliability
If data migration is performed proactively based on read-count tables, then read disturbance is avoided, but additional write operations and processing overhead are introduced
Solution Approach 1:
The system employs self-service by automatically monitoring its own read patterns through read-count tables and autonomously deciding when data migration is necessary. The flash memory controller itself generates the feedback information and executes migration operations without external intervention, reducing the need for complex external management systems.
Solution Approach 2:
The system changes parameters by tracking read counts and using these quantitative measurements to determine migration timing. By converting qualitative read patterns into quantitative read-count data, the system can objectively identify when migration is needed, simplifying the decision-making process compared to qualitative assessment methods.
3Loss of information
If read-count tables are maintained to predict singular blocks, then data loss is prevented, but additional memory and processing resources are consumed
Solution Approach 1:
The system extracts only the essential information needed for prediction by maintaining read-count tables that track specific read operations. Rather than monitoring all possible block parameters, the system focuses on extracting and storing only the read count data necessary for predicting singular block formation, minimizing resource consumption while maintaining effectiveness.
Data Source
AI summary
A method for migrating data to avoid read disturbance is introduced to contain the following steps: obtaining first read counts corresponding to physical blocks from a first read-count table; obtaining second read counts corresponding to the physical blocks from a second read-count table; subtracting each first read count from one corresponding second read count to generate third read counts; finding a singular physical-block from physical blocks according to the third read counts; performing a test read on data of the ith physical page of the singular physical-block; determining whether the data of the ith physical page of the singular physical block has passed the test read; and if so, moving or copying data of the ith physical page and at least one neighboring physical-page of the singular physical block to an available physical-block.