In-Situ Data Refresh Using Time-Varying ECC for Memory Retention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in maintaining a reliable read window budget (RWB) due to phenomena like Inter-cell Interference, Cell-to-Cell coupling, and Lateral Charge Loss, leading to data degradation and reduced data retention, which conventional error correction methods struggle to address effectively without compromising storage capacity or frequency of operations.
Innovation Solution
Implementing a data refresh mechanism that varies the frequency of error correction operations based on experimentally known data degradation rates, using multiple portions of error-correction data stored in different locations, and periodically deleting used portions to maintain a target RWB while optimizing storage capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional error correction methods are used to address data degradation, then data retention is improved, but storage capacity is reduced
Solution Approach 1:
The error correction data is divided into multiple portions stored at different locations within the memory device. This segmentation allows the system to perform error correction operations on specific portions without requiring all data to be stored separately, thereby maintaining storage capacity while improving data retention through targeted refresh operations.
Solution Approach 2:
The system performs error correction operations periodically based on experimentally known data degradation rates. By timing the refresh operations to occur at optimal intervals rather than continuously, the system maintains data retention without the overhead of constant operations, thus preserving storage capacity for user data.
2Reliability
If frequent error correction operations are performed to maintain data retention, then data reliability is improved, but storage capacity is reduced
Solution Approach 1:
The system dynamically adjusts the frequency and timing of error correction operations based on experimentally determined data degradation rates. This dynamic approach allows the system to perform corrections only when necessary, optimizing the balance between data reliability and storage capacity utilization.
Solution Approach 2:
The system changes the parameter of operation frequency based on data degradation characteristics. By adjusting when error correction operations occur rather than executing them at fixed intervals or continuously, the system maintains reliability while maximizing available storage capacity for user data.
3Reliability
If error correction operations are performed to maintain target RWB levels, then data retention is improved, but bandwidth and power consumption increase
Solution Approach 1:
The system performs error correction operations periodically rather than continuously, timing them to occur at intervals determined by data degradation rates. This periodic approach maintains target RWB levels while significantly reducing power consumption compared to continuous operations.
Solution Approach 2:
The system uses experimentally known data degradation rates to autonomously determine optimal refresh timing without requiring constant monitoring or external intervention. This self-service approach maintains data retention while minimizing power consumption by performing operations only when degradation necessitates correction.
Data Source
AI summary
A system can include a memory device and a processing device, operatively coupled with the memory device, to perform operations including storing a set of user data and multiple portions of error correction data. The operations can also include, responsive to an expiration of a first threshold amount of time after storing the set of user data, performing, using the third portion of the error correction data, a first error correction operation, on each of the set of user data, the first portion, and the second portion, and rewriting, on the memory device, the set of user data, the first portion, and the second portion. The operations can further include deleting the third portion.