Dynamic Patrol Frequency for NAND Flash Wear Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The data retention capability of NAND flash memory in semiconductor memory systems decreases over time due to wear, leading to potential data corruption and errors, which existing patrol processing methods fail to address effectively due to fixed and non-optimal examination frequencies.
Innovation Solution
A memory system comprising a nonvolatile semiconductor memory, a temperature sensor, and a controller that dynamically sets the examination frequency for data quality based on wear levels and usage patterns, using a patrol management table to prioritize patrol processing for blocks with higher wear levels, thereby optimizing the execution cycle of data quality examination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed patrol processing frequency is used, then implementation is simple, but data retention capability deteriorates due to insufficient examination of high-wear blocks
Solution Approach 1:
The patent implements dynamic patrol processing frequency adjustment based on wear levels. The controller examines wear levels of memory blocks and adjusts the patrol frequency accordingly - high-wear blocks receive more frequent examination while low-wear blocks are examined less frequently. This dynamic adaptation resolves the contradiction by maintaining simple implementation while improving data retention capability through intelligent frequency modulation.
Solution Approach 2:
The patent applies different patrol processing frequencies to different memory blocks based on their individual wear levels. Instead of uniform treatment, the system identifies blocks with higher wear levels and prioritizes them for examination. This local differentiation ensures that resources are allocated efficiently to where they are most needed, improving overall data retention without unnecessarily increasing system complexity.
2Reliability
If patrol processing frequency is increased, then data quality examination is improved, but user data processing performance deteriorates due to interruptions
Solution Approach 1:
The system dynamically adjusts patrol frequency based on actual wear conditions rather than using a fixed high frequency. By monitoring wear levels and adapting the examination frequency accordingly, the system maintains adequate data quality checking while minimizing interruptions to user data processing operations. This dynamic modulation allows the system to achieve both reliability and productivity goals.
Solution Approach 2:
The patent changes the patrol processing parameter (frequency) based on wear level conditions. The controller modifies the examination frequency as a variable parameter rather than maintaining a constant value. This parameter adaptation enables the system to achieve sufficient data quality examination when needed while reducing processing interruptions during normal operation, thus resolving the contradiction between reliability and productivity.
3Device complexity
If uniform patrol frequency is applied to all blocks, then system simplicity is maintained, but wear level minimization is ineffective for high-wear blocks
Solution Approach 1:
The patent implements local quality differentiation by applying distinct patrol frequencies to different blocks based on their wear levels. The controller identifies blocks with higher wear levels and assigns them higher examination frequencies, while blocks with lower wear levels receive reduced frequency. This localized approach minimizes overall wear by preventing unnecessary examinations of low-wear blocks while ensuring adequate protection for high-wear blocks, all within a manageable system structure.
Solution Approach 2:
The system segments the memory blocks into categories based on wear levels (e.g., high-wear, medium-wear, low-wear blocks). This segmentation allows the controller to apply different patrol strategies to different segments. By dividing the memory space into wear-based groups, the system achieves effective wear minimization without requiring complete redesign of the patrol management structure, thus resolving the contradiction between simplicity and effectiveness.
Data Source
AI summary
According to one embodiment, a memory system comprises a first nonvolatile semiconductor memory, a temperature sensor and a controller. The first nonvolatile semiconductor memory includes the first and second semiconductor chips. The temperature sensor detects a temperature of the first nonvolatile semiconductor memory. The controller acquires the wear level per block of the first and second semiconductor chips based on the temperature of the first nonvolatile semiconductor memory and the frequency of use of the first nonvolatile semiconductor memory, and sets, based on the wear level, an examination frequency for defining a cycle of examination of quality of data per block of the first and second semiconductor chips.