Controller Selects Target Open Regions for Time-Distributed Refresh
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data retention characteristics in semiconductor memory cells are affected by factors such as access frequency and interference between memory cells, leading to data loss or deformation, reducing the reliability of memory apparatuses.
Innovation Solution
A data storage device with a controller that selects target open regions based on health information, including read count and elapsed time, to perform time-distributed refresh operations, minimizing disturbance vulnerability and maintaining data integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If refresh operations are performed frequently on all open regions, then data retention reliability is improved, but read disturb vulnerability and energy consumption increase
Solution Approach 1:
The patent applies local quality by differentiating refresh strategies based on the specific characteristics of each open region. Instead of uniformly refreshing all open regions, the controller identifies target open regions with higher data loss risk (those with longer elapsed times since last program operation) and applies refresh operations selectively to these regions, while avoiding unnecessary refreshes in regions with lower risk.
Solution Approach 2:
The patent implements partial action by performing refresh operations on only a subset of open regions (target open regions) rather than all open regions. This selective approach reduces the total number of refresh operations, thereby minimizing read disturb vulnerability and energy consumption while still maintaining adequate data retention for the most vulnerable regions.
2Reliability
If refresh operations are performed on all open regions simultaneously, then data retention is improved, but system productivity and access speed deteriorate due to time-consuming operations
Solution Approach 1:
The patent segments the refresh operation into two distinct phases: a first refresh operation performed immediately on target open regions to prevent data loss, and a second refresh operation performed later on non-target open regions. This segmentation allows the system to prioritize critical refresh operations while deferring less urgent ones, thereby maintaining data retention without completely blocking system access.
Solution Approach 2:
The patent implements periodic action by scheduling refresh operations at different time points. The first refresh operation is performed immediately when open regions are identified, while the second refresh operation is performed after a predetermined period or when system conditions allow. This periodic approach ensures data retention while minimizing impact on system productivity.
3Loss of energy
If selective refresh operations are performed only on target open regions, then energy consumption is reduced, but data retention reliability may worsen if not properly managed
Solution Approach 1:
The patent applies preliminary action by performing a first refresh operation immediately on target open regions before data loss can occur. This early intervention ensures that the most vulnerable data (in open regions with longer elapsed times) is refreshed first, establishing a safety net against data loss while avoiding the need to continuously refresh all open regions.
Solution Approach 2:
The patent implements feedback by monitoring the elapsed time since last program operation for each open region and using this information to identify target open regions that require immediate refresh. This feedback mechanism ensures that refresh operations are directed to regions where they are most needed, balancing energy consumption with data retention reliability.
Data Source
AI summary
A data storage device may include a storage including a plurality of storage regions each composed of a plurality of pages; and a controller. The controller is configured to select a plurality of target open regions from open regions among the storage regions on the basis of health information of the open regions, in each of which a programmed page and an unprogrammed page coexist, and perform control so that refresh operations for the respective target open regions are performed in a time-distributed manner.