Memory Controller Victim Block Selection for Error Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in managing and preventing errors during data programming, leading to potential data loss and reduced operational efficiency due to the lack of effective error detection and correction mechanisms, especially in memory blocks with high program error rates and prolonged read operations.
Innovation Solution
A memory system with a controller that detects program error bit information, selects a 'victim' memory block based on error rates and read operation metrics, and copies data from this block to a free block without performing error correction, using a garbage collection operation to ensure data validity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If program verify operation is performed to detect errors, then data reliability is improved, but operation time increases
Solution Approach 1:
The patent performs program verify operation during the programming process to detect errors early, before data is fully written to the memory block. This preliminary verification allows the system to identify victim blocks proactively, preventing future read errors without waiting for actual read operations to fail.
Solution Approach 2:
The patent creates a cushioning mechanism by maintaining an error bit count for each memory block and comparing it against a threshold. This beforehand cushioning allows the system to prepare for potential failures by identifying and copying data from at-risk blocks before actual errors occur during read operations, thus preventing data loss without continuous verification.
2Reliability
If data is copied from victim memory blocks to prevent errors, then data stability is improved, but device complexity increases
Solution Approach 1:
The memory system performs self-service by automatically monitoring its own health through program verify operations, identifying problematic blocks, and copying data from these blocks without external intervention. The controller autonomously manages error detection, victim block identification, and data migration, reducing the need for complex external error correction mechanisms.
Solution Approach 2:
Instead of implementing complex error correction codes or redundant storage systems, the patent uses a simple copying mechanism. When a memory block is identified as a victim block through error detection during programming, the system copies its contents to a new memory block, effectively replacing corrupted data with fresh copies without requiring sophisticated error correction algorithms.
3Measurement precision
If program verify operation is performed, then error detection capability is improved, but productivity decreases
Solution Approach 1:
The patent applies program verify operation selectively rather than universally. Instead of verifying every memory block after programming, the system focuses verification efforts on blocks that are more prone to errors or have shown previous issues. This localized approach maintains high error detection capability while reducing the overall verification burden and preserving productivity.
Data Source
AI summary
A memory system comprises a memory device including a plurality of memory blocks, the memory device being configured to perform a program operation and a program verify operation to program data to the memory blocks, and a controller configured to detect program error bit information as a result of the program verify operation, select a victim memory block among the memory blocks based on the detected program error bit information, and copy programmed data of the victim memory block.