Memory Subsystem Powered-Off Duration Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sub-systems face challenges in accurately determining the powered-off duration, which can lead to inadequate data refresh operations and increased read-error rates due to charge loss over time.
Innovation Solution
The solution involves using a dedicated reference block to measure time-based changes in memory cell characteristics, allowing the memory sub-system controller to estimate the powered-off duration without relying on external indicators, and to initiate data refresh operations such as folding to maintain data integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the memory sub-system relies on external indicators to determine powered-off duration, then the measurement may be more accurate, but the system complexity and dependency on external components increases
Solution Approach 1:
The memory sub-system performs self-measurement of the powered-off duration by detecting changes in memory cell characteristics (such as threshold voltage drift or charge loss) that occur naturally over time when powered off. This eliminates the need for external time-indicating components, as the memory system uses its own internal physical changes as the measurement basis, thereby reducing system complexity while maintaining measurement capability.
Solution Approach 2:
The patent introduces an intermediary reference block or reference memory cells that are specifically designed to exhibit predictable time-dependent characteristic changes. These reference elements serve as mediators between the passage of time and the measurement system, allowing the controller to infer powered-off duration by comparing the reference block's characteristics against known degradation patterns, thus achieving accurate measurement without direct external timekeeping.
2Loss of time
If data refresh operations are delayed, then system latency is reduced, but read-error rates increase due to charge loss
Solution Approach 1:
The system performs preliminary assessment of data integrity by detecting early signs of charge loss or threshold voltage drift in the reference block during or immediately after power-on. By identifying the powered-off duration through these preliminary detections, the system can proactively determine whether refresh operations are needed before actual data corruption occurs, thus balancing latency and reliability by refreshing only when necessary.
Solution Approach 2:
The patent implements a feedback mechanism where the measured powered-off duration (derived from memory cell characteristic changes) continuously informs the refresh operation scheduling. The controller uses this feedback to dynamically adjust refresh timing, performing refresh operations only when the measured duration exceeds a threshold that indicates impending data corruption, thereby minimizing unnecessary refresh operations and system latency while maintaining data reliability.
3Reliability
If the memory sub-system performs frequent data refresh operations, then data integrity is maintained, but system performance and productivity decrease
Solution Approach 1:
Instead of performing exhaustive refresh operations on all memory blocks regardless of need, the patent applies partial action by selectively refreshing only those blocks or regions that exhibit signs of charge loss or have been powered off for extended durations. This targeted approach maintains data integrity for at-risk data while avoiding unnecessary refresh operations on stable data, thus preserving system performance while ensuring reliability.
Solution Approach 2:
The system dynamically adjusts refresh operation parameters (such as refresh frequency, scope, and intensity) based on the measured powered-off duration and detected memory cell characteristics. When the powered-off duration is short and characteristics are stable, refresh operations are minimized or skipped. When duration exceeds thresholds or degradation is detected, refresh parameters are adjusted to ensure data integrity. This adaptive parameter adjustment optimizes the balance between reliability and productivity.
Data Source
AI summary
A processing device in a memory sub-system detects a power-on command for the memory device. The processing device obtains read information from a reference block of the memory device. The read information is based on reference data stored in the reference block. Based on the read information, the processing device determines an estimate of a duration for which the memory device was in a powered-off state. The processing device determines whether the duration satisfies a duration threshold, and responsive to determining the duration satisfies the duration threshold, the processing device initiates a folding operation for at least a subset of blocks of the memory device.