Adaptive Data Integrity Scan Frequency for Memory Subsystems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory subsystems face inefficiencies in data integrity scans due to fixed scan windows, leading to increased bandwidth usage and reduced reliability in areas with better-than-worst-case memory performance, as they scan more frequently than necessary, potentially causing read disturb errors.
Innovation Solution
Implementing an adaptive data integrity scan scheme that dynamically adjusts the scan frequency based on actual reliability metrics, using a scaling factor and hysteresis to optimize scan windows, reducing unnecessary scans in reliable areas while maintaining coverage for less reliable ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed scan window is used for data integrity scans, then worst-case reliability scenarios are covered, but bandwidth is increased and scan frequency is reduced in areas with better-than-worst-case performance
Solution Approach 1:
The patent applies local quality by dividing the memory space into multiple regions and assigning different scan window sizes to different regions based on their individual reliability characteristics. High-reliability regions receive smaller scan windows while low-reliability regions receive larger scan windows, optimizing bandwidth usage while maintaining appropriate reliability coverage for each region.
Solution Approach 2:
The patent implements dynamics by making the scan window size adaptive rather than fixed. The system dynamically adjusts the scan window size based on measured reliability metrics, allowing the scan frequency and window size to change over time according to actual memory performance rather than using a static worst-case configuration throughout.
2Reliability
If scan frequency is increased to ensure data integrity, then reliability is improved, but bandwidth is consumed and read disturb errors may increase
Solution Approach 1:
The patent applies local quality by implementing region-specific scan frequencies based on local reliability characteristics. Instead of uniformly increasing scan frequency across the entire memory space, the system identifies specific regions that require more frequent scanning and applies larger scan windows only to those regions, thereby maintaining data integrity where needed while minimizing read disturb in high-reliability regions.
Solution Approach 2:
The patent implements feedback by continuously monitoring reliability metrics and using this information to adjust scan window sizes and frequencies. The system measures actual memory reliability performance and feeds this information back to the scan management logic, which then adapts the scanning strategy to match actual conditions rather than relying on static worst-case assumptions.
3Loss of energy
If a probabilistic data integrity scan scheme is implemented, then bandwidth usage is reduced, but scan frequency must be optimized to maintain reliability coverage
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the scan window size parameter based on measured reliability metrics. The system changes the key parameter (scan window size) of the probabilistic scan scheme to optimize the balance between bandwidth usage and reliability coverage, allowing the probabilistic approach to work effectively with adaptive parameters rather than fixed values.
Data Source
AI summary
Exemplary methods, apparatuses, and systems include receiving a plurality of read operations. The read operations are divided into a current set of a sequence of read operations and one or more other sets. The size of the current set is a first number of read operations. An aggressor read operation is selected from the current set. A first data integrity scan is performed on a victim of the aggressor and a first indicator of data integrity is determined based on the first data integrity scan. A scaling factor is determined using the indicator of data integrity and a number of program erase cycles for the portion of memory. The set size of read operations is adjusted to a second number of read operations using the scaling factor for a subsequent set.