Memory Page Integrity Checks Using Voltage Distribution Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data integrity checks in memory sub-systems based on single metrics like RBER or BEC may lead to excessive or insufficient refreshing of memory blocks, causing resource inefficiency and reliability issues due to increased error rates.
Innovation Solution
Implementing a data integrity check using programming distribution valley metrics, including valley margin, valley floor, and valley center, to determine when media management operations like block refreshing are necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If data integrity checks use single metrics like RBER or BEC, then the check process is simple, but it causes excessive or insufficient refreshing leading to resource inefficiency and reliability issues
Solution Approach 1:
The patent segments the data integrity check into multiple independent metrics (valley margin, valley floor, valley center) instead of using a single metric. Each metric provides a specific aspect of data integrity information, allowing the system to comprehensively assess block health without oversimplifying the evaluation process.
Solution Approach 2:
The patent transitions from one-dimensional metrics (RBER, BEC) to three-dimensional metrics by introducing valley margin, valley floor, and valley center. This dimensional expansion enables a more comprehensive assessment of data integrity, capturing multiple aspects of programming distribution simultaneously.
2Reliability
If data integrity checks use multiple metrics like valley margin, valley floor, and valley center, then data reliability assessment improves, but the check process becomes more complex
Solution Approach 1:
The patent merges multiple metrics (valley margin, valley floor, valley center) into a unified data integrity check framework. By combining these metrics, the system achieves comprehensive reliability assessment while managing complexity through integration rather than separate independent checks.
Solution Approach 2:
The multi-metric framework serves multiple functions: it assesses different aspects of data integrity simultaneously (programming distribution quality, threshold accuracy, error rates). This universal approach allows a single check process to address various reliability concerns without requiring separate specialized checks for each metric.
3Speed
If single metrics like RBER or BEC are used for refreshing decisions, then the decision process is fast, but it leads to excessive refreshing causing resource waste
Solution Approach 1:
The patent implements a feedback mechanism where multiple metrics (valley margin, valley floor, valley center) provide comprehensive information about block health status. This feedback allows the system to make more accurate refreshing decisions, avoiding unnecessary operations while ensuring data integrity, thereby reducing energy consumption.
Solution Approach 2:
The patent changes the parameters used for refreshing decisions from single metrics (RBER, BEC) to multiple metrics (valley margin, valley floor, valley center). This parameter change enables more precise assessment of when refreshing is actually needed, reducing excessive refreshing operations and associated energy waste.
4Reliability
If frequent refreshing operations are performed to ensure data integrity, then reliability improves, but productivity decreases due to resource consumption
Solution Approach 1:
The patent applies partial action by using multiple metrics to assess data integrity before initiating refreshing operations. Instead of performing frequent comprehensive checks and potential refreshing, the system uses the multi-metric framework to determine when refreshing is actually necessary, reducing unnecessary operations and maintaining productivity.
Solution Approach 2:
The patent performs preliminary assessment using multiple metrics (valley margin, valley floor, valley center) before executing refreshing operations. This preliminary action allows the system to evaluate block health comprehensively and make informed decisions about whether refreshing is needed, avoiding premature or unnecessary refreshing that would reduce productivity.
Data Source
AI summary
Systems and methods are disclosed including a memory device and a processing device operatively coupled to the memory device. The processing device can perform operations including determining a value of a data state metric of a memory page, wherein the data state metric value is reflective of a number of bit errors associated with the memory page; upon determining that the data state metric value satisfies a first threshold criterion, determining at least two different voltage distribution metrics values currently associated with the memory page, wherein each voltage distribution metrics value reflects one of a voltage distribution margin, a voltage distribution floor, or a voltage distribution center; and responsive to one or more of the voltage distribution metrics values satisfying a second threshold criterion, performing a media management operation with respect to a block associated with the memory page.