Variable Read Reclaim for Super Block Page Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In memory systems using nonvolatile memory devices, read disturbance leads to increased error bits in frequently read memory blocks, necessitating a read reclaim operation to move data before uncorrectable errors occur, but existing methods perform this operation at the super block level, potentially increasing unnecessary erase/write cycles in blocks with lower read counts.
Innovation Solution
A memory system with a controller that configures memory blocks into super blocks, divides them into page groups, and performs read reclaim operations based on memory-block-read-counts, selectively targeting page groups with higher read counts to prevent data loss and optimize erase/write cycles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If read reclaim operation is performed at the super block level, then data loss is prevented, but unnecessary erase/write cycles increase in blocks with lower read counts
Solution Approach 1:
The super block is divided into multiple page groups, allowing the read reclaim operation to be performed selectively on specific page groups rather than the entire super block. This segmentation enables targeted intervention on high-read-count page groups while preserving low-read-count page groups from unnecessary erase/write cycles.
Solution Approach 2:
Different page groups within the super block are treated differently based on their individual read counts. Page groups with read counts above a threshold undergo read reclaim operations, while those below the threshold are preserved. This local differentiation optimizes the balance between data protection and resource consumption.
2Reliability
If read reclaim operation is performed on all page groups, then data loss is prevented, but operation efficiency decreases due to unnecessary operations
Solution Approach 1:
Instead of performing read reclaim operations on all page groups (excessive action), the system performs operations only on the necessary subset of page groups with high read counts (partial action). This selective approach maintains data protection for vulnerable data while avoiding wasteful operations on already-safe data.
Solution Approach 2:
The system uses read count as a dynamic parameter to determine which page groups require read reclaim operations. By monitoring and comparing read counts against a threshold, the system adaptively adjusts the scope of read reclaim operations, improving efficiency while maintaining reliability.
Data Source
AI summary
A memory system may include a plurality of nonvolatile memory devices each including a plurality of memory blocks; and a controller configured to configure at least one memory block selected from the plurality of memory blocks as a super block, and to perform a read reclaim operation for the super block based on memory-block-read-counts of the super block, wherein the controller divides the super block into a plurality of page groups, selects at least one page group from the plurality of page groups based on the memory-block-read-counts, and performs the read reclaim operation.