Word Line Group Read Counters for Memory Disturb Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional read-disturb management techniques in memory sub-systems rely on per block read counters, leading to inefficient utilization of memory devices as they result in premature folding of numerous cells, as they track read operations at a block level rather than a word line group level.
Innovation Solution
Maintaining read counters on a per word line group basis to track read operations and perform read-disturb handling, allowing for more precise monitoring and handling of read operations, thereby avoiding premature folding of cells and improving memory resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If per block read counters are used to track read operations, then read-disturb management can be implemented, but memory resource utilization becomes inefficient due to premature folding of cells
Solution Approach 1:
The patent divides the memory block into multiple word line groups and implements separate read counters for each word line group instead of using a single counter per block. This segmentation allows independent tracking of read operations at the word line group level, enabling more granular read-disturb management and preventing premature folding of entire blocks when only specific word line groups have reached their read limits.
2Reliability
If read operations are tracked at block level, then read-disturb handling can be performed, but premature folding of cells occurs reducing operational lifespan
Solution Approach 1:
The patent segments the block-level tracking into word line group-level tracking by maintaining separate read counters for each word line group. This enables the system to identify and handle read-disturb conditions at the precise location (specific word line group) rather than forcing folding of entire blocks, thereby extending the operational lifespan of memory cells by avoiding unnecessary folding operations.
Solution Approach 2:
The patent applies local quality by implementing read-disturb management with different thresholds and handling strategies for different word line groups based on their specific read operation patterns. Each word line group can have its own read-disturb threshold, allowing the system to optimize read-disturb handling locally rather than applying a uniform block-level threshold, thus preventing premature folding and extending cell lifespan.
Data Source
AI summary
A processing device detects a read operation at a memory device that is directed at a word line group from among multiple word line groups of the memory device. The processing device increments a read counter associated with the word line group based on the read operation being directed at the word line group. The processing device determines the read counter exceeds a read-disturb threshold and performs read-disturb handling on the word line group in response to determining the read counter exceeds the read-disturb threshold.