Probabilistic Data Integrity Scans for Localized Read Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory subsystems face issues with read disturb errors due to repetitive data integrity scans triggered by localized read patterns, leading to unnecessary resource consumption and potential quality of service compromises.
Innovation Solution
Implementing a probabilistic data integrity scheme that caches aggressor read operations and uses counters to manage the frequency of data integrity scans, thereby reducing unnecessary checks on victim cells and mitigating read disturb effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data integrity scans are performed frequently to detect read disturb errors, then data reliability is improved, but resource consumption increases and quality of service deteriorates
Solution Approach 1:
The patent applies local quality by differentiating between frequently accessed memory locations (aggressors) and infrequently accessed locations. Data integrity scans are selectively performed only on victim cells adjacent to aggressors with localized read patterns, rather than uniformly scanning all memory. This targeted approach maintains data reliability for vulnerable regions while reducing overall scan frequency and resource consumption.
Solution Approach 2:
The patent implements partial action by performing data integrity scans on only a subset of memory cells - specifically, victim cells adjacent to aggressors exhibiting localized read patterns. Instead of scanning all memory cells uniformly, the system performs scans only where read disturb errors are most likely to occur, based on observed access patterns. This reduces the total number of scans while maintaining adequate error detection coverage.
2Reliability
If data integrity scans are performed frequently to ensure data reliability, then read disturb errors are detected, but resource consumption increases unnecessarily
Solution Approach 1:
The patent implements partial action by performing data integrity scans on only a subset of memory cells - specifically, victim cells adjacent to aggressors exhibiting localized read patterns. Instead of scanning all memory cells uniformly, the system performs scans only where read disturb errors are most likely to occur, based on observed access patterns. This reduces the total number of scans while maintaining adequate error detection coverage.
Solution Approach 2:
The patent applies periodic action by monitoring read patterns over time and adjusting scan frequency dynamically. Rather than performing scans at fixed intervals or continuously, the system periodically evaluates whether localized read patterns are present and only intensifies scanning when such patterns are detected. This adaptive periodic approach reduces average resource consumption while maintaining reliability when needed.
Data Source
AI summary
Exemplary methods, apparatuses, and systems include receiving read operations. The read operations are divided into a current set of a sequence of read operations and one or more other sets of sequences of read operations. An aggressor read operation is selected from the current set. A data integrity scan is performed on a victim of the aggressor read operation. A read margin for a victim of the aggressor read operation is determined based on the error rate. An identifier associated with the aggressor is added to a cache and a counter for the identifier added to the cache is initialized based upon the determined read margin.