Read Disturb Mitigation in Non-Volatile Memory Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor memory systems face challenges with read disturb, a phenomenon that causes data integrity issues due to frequent reading operations, leading to failed bits and reduced memory lifespan, especially in high read-intensive blocks.
Innovation Solution
Implementing a write scheme that reduces read disturb by storing dummy data in certain memory cells and strategically placing user/system data on alternating or specific word lines, and using a relaxed read scrub criterion for high read-intensive blocks to minimize the frequency of read scrub operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If frequent read operations are performed on memory blocks, then data accessibility is improved, but read disturb increases causing data integrity degradation
Solution Approach 1:
The patent applies preliminary action by performing read scrub operations proactively before data integrity is completely compromised. The system periodically reads data from memory blocks, corrects any read disturb-induced errors using error correction codes, and rewrites the corrected data back to the same or different blocks, thereby preventing data corruption before it affects reliability.
Solution Approach 2:
The patent implements discarding and recovering by temporarily moving data out of high read-intensive blocks during read scrub operations. Data is read from the original block, corrected if necessary, and then rewritten either to the same block (after verification) or to a different block, effectively discarding the corrupted data state and recovering the correct data state.
2Reliability
If read scrub operations are performed frequently to correct failed bits, then data integrity is maintained, but memory cell wear increases reducing lifespan
Solution Approach 1:
The patent applies partial action by performing read scrub operations selectively rather than on all memory blocks uniformly. The system identifies high read-intensive blocks that are more susceptible to read disturb and targets these blocks for read scrub, while reducing or skipping read scrub on low read-intensive blocks, thereby maintaining data integrity where needed while minimizing unnecessary wear on less vulnerable blocks.
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting the read scrub frequency and criteria based on memory block characteristics. The system monitors read intensities, error rates, and block ages to adaptively determine which blocks require read scrub and how frequently, changing the operational parameters of read scrub from a fixed uniform approach to a dynamic selective approach that extends memory lifespan.
3Reliability
If data is rewritten to different locations during read scrub, then data integrity is protected, but device complexity increases
Solution Approach 1:
The patent applies universality by designing the read scrub system to handle multiple functions within a unified framework. The same read scrub mechanism handles both error correction and data relocation, the error correction code system handles both detection and correction, and the memory management system handles both reading and rewriting operations, thereby protecting data integrity without proportionally increasing device complexity.
Data Source
AI summary
A non-volatile storage system comprises non-volatile memory cells arranged in physical blocks, and one or more control circuits in communication with the non-volatile memory cells. The one or more control circuits are configured to write data to a physical block of the non-volatile memory cells with a scheme to reduce read disturb if a logical block associated with the physical block has a read intensity greater than a threshold.


