Adaptive Read Disturb Scanning for Asymmetric Memory Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional read disturb scans fail to effectively address read disturb errors in asymmetric blocks, leading to uneven stress distribution and increased bit-error rates, particularly in partially filled blocks, resulting in high scan under-queue rates and unrecoverable error correcting code data loss.
Innovation Solution
An adaptive read disturb scan is implemented to identify and target high-risk sections within asymmetric blocks, executing scans on unprogrammed wordlines of identified sections to mitigate uneven stress and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional read disturb scan is performed on asymmetric blocks, then scan coverage is provided, but uneven stress distribution occurs leading to increased bit-error rates
Solution Approach 1:
The patent applies local quality by performing read disturb scans selectively on specific sections of asymmetric blocks rather than uniformly across all blocks. The controller identifies high-risk sections within asymmetric blocks and concentrates scan operations on those sections, creating non-uniform stress distribution that targets vulnerable areas while preserving overall reliability.
2Reliability
If read disturb scan is performed on all blocks, then comprehensive error detection is achieved, but latency increases due to unnecessary scans on low-risk blocks
Solution Approach 1:
The patent implements partial action by performing read disturb scans only on identified high-risk sections of asymmetric blocks rather than on all blocks. The controller uses block metadata and asymmetry information to determine which sections require scanning, applying scan operations partially where needed and skipping low-risk areas, thereby reducing overall scan latency while maintaining adequate error detection capability.
3Device complexity
If asymmetric blocks are handled with conventional scan methods, then device complexity is minimized, but scan under-queue rates increase leading to data loss
Solution Approach 1:
The patent applies segmentation by dividing asymmetric blocks into distinct sections and treating each section differently based on its risk profile. The controller segments the block management process into identification of asymmetric blocks, determination of high-risk sections, and targeted scan execution on those sections. This segmented approach handles device complexity through systematic classification while improving data integrity through focused error prevention.
Data Source
AI summary
A system comprises a memory device and a processing device, operatively coupled to the memory device. The processing device receives a request to perform a read disturb scan for a block of the memory device, wherein the block is partially programmed. The processing device determines the block is asymmetric, wherein the block comprises a plurality of sections, and wherein each section comprises memory cells corresponding to a different physical block in the memory device. Responsive to determining the block is asymmetric, the processing device identifies a target section of the block. The processing device executes the read disturb scan on an unprogrammed wordline of the target section.