Nonvolatile Memory Subblock Reclaim for Read Disturb
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing nonvolatile memory devices face challenges in efficiently managing and maintaining data integrity due to read disturb and program/erase cycles, leading to reduced reliability and operational speed, especially in multi-subblock memory structures.
Innovation Solution
A control method for nonvolatile memory devices that includes comparing count values of program, read, and erase operations with a reference value, performing refresh or reclaim operations when the count exceeds the threshold, and strategically managing data between subblocks to mitigate read disturb and maintain data stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple sub-blocks within a memory block are erased independently, then erase flexibility and data management efficiency are improved, but read disturb effects across sub-blocks increase leading to potential data loss
Solution Approach 1:
The patent applies preliminary action by performing reclaim operations on sub-blocks before they accumulate excessive read disturb effects. The controller monitors operation counts and proactively transfers data from sub-blocks that approach the reference threshold, preventing read disturb from reaching levels that would cause data loss. This proactive approach maintains data integrity while enabling independent erase operations.
2Adaptability or versatility
If the memory device allows independent erase operations on each sub-block, then operational flexibility is improved, but the complexity of managing and monitoring multiple sub-blocks increases
Solution Approach 1:
The patent implements self-service by enabling the memory device to autonomously monitor operation counts and manage reclaim operations without external intervention. The controller automatically compares operation counts against reference values and initiates reclaim operations when thresholds are met, reducing the need for complex external management while maintaining erase flexibility.
Solution Approach 2:
The patent employs feedback mechanisms where the controller continuously monitors operation counts of sub-blocks and uses this information to determine when reclaim operations are necessary. This feedback loop enables the system to adapt its behavior based on actual usage patterns, maintaining flexibility while simplifying management through automated decision-making.
3Reliability
If data is transferred between sub-blocks to prevent read disturb, then data integrity is improved, but additional operations and time consumption increase
Solution Approach 1:
The patent applies preliminary action by performing data transfer operations before read disturb accumulates to problematic levels. By monitoring operation counts and triggering reclaim operations at predetermined thresholds, the system prevents the need for more extensive data recovery operations later, thereby reducing overall time loss while maintaining integrity.
Solution Approach 2:
The patent uses partial action by transferring data only from sub-blocks that approach the reference operation count threshold, rather than performing comprehensive operations on all sub-blocks. This selective approach minimizes the time and resources required for data management while still preventing read disturb-related data loss.
Data Source
AI summary
According to example embodiments, a control method of a nonvolatile memory device, which includes a plurality of memory blocks on a substrate, each memory block including a plurality of sub blocks stacked in a direction perpendicular to the substrate and being configured to be erased independently and each sub block including a plurality of memory cells stacked in the direction perpendicular to the substrate. The control method includes comparing a count value of a first memory block with a reference value, the count value determined according to the number of program, read, or erase operations executed at the first memory block after data is programmed in the first memory block; and if the count value is greater than or equal to the reference value, performing a reprogram operation in which data programmed in first the memory block is read and the read data is programmed in a second memory block.