Memory Management Method for Wear Leveling in Non-Volatile Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rewritable non-volatile memory modules experience uneven wear across devices due to differing usage frequencies, leading to potential damage and reduced lifespan, as frequently updated file system data increases wear values of specific devices more rapidly than others.
Innovation Solution
A memory management method that configures logical units into groups and maps them to physical erasing units across multiple devices, performing remapping operations when wear values become uneven, ensuring data is written to devices with lower wear values to distribute usage evenly and maintain balanced wear levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If file system data is frequently updated, then data accessibility and system performance are improved, but wear value of physical erasing units increases rapidly
Solution Approach 1:
The patent divides the memory storage apparatus into multiple devices (first device, second device, etc.), each with its own physical erasing units. By segmenting the storage capacity across multiple devices, the system can distribute the wear burden and prevent any single physical erasing unit from experiencing excessive write cycles, thus resolving the contradiction between frequent data updates and wear accumulation.
Solution Approach 2:
The patent implements dynamic remapping of logical units to physical erasing units based on wear values. The memory control circuit continuously monitors wear levels and adjusts the mapping relationships, allowing the system to adaptively redirect frequently accessed data to less worn devices. This dynamic adjustment enables the system to maintain high data accessibility while balancing wear distribution across all physical erasing units.
2Productivity
If wear values of devices are not leveled, then device performance remains high, but overall system reliability decreases due to excessive wear on specific devices
Solution Approach 1:
The patent employs a feedback mechanism where the memory control circuit monitors the wear values of multiple devices and uses this information to adjust the mapping between logical units and physical erasing units. When a device's wear value exceeds a threshold or becomes significantly higher than other devices, the system automatically remaps data to balance the wear distribution. This feedback loop ensures that all devices operate within reliable wear limits while maintaining overall system performance.
Solution Approach 2:
The patent changes the parameter of mapping relationships between logical units and physical erasing units based on wear values. By dynamically adjusting these mapping parameters, the system can redistribute data access patterns to balance wear across devices. This parameter change strategy allows the system to maintain high performance by ensuring no single device becomes a bottleneck due to excessive wear.
3Reliability
If remapping operation is performed frequently, then wear levels are balanced effectively, but system complexity and overhead increase
Solution Approach 1:
The patent performs preliminary actions by pre-establishing multiple mapping relationships between logical units and physical erasing units before wear imbalance occurs. The system maintains a pool of available devices and their corresponding physical erasing units, ready to be remapped when needed. This preparation reduces the complexity of actual remapping operations, as the system only needs to switch between pre-configured mappings rather than creating new relationships on-the-fly.
Data Source
AI summary
A memory management method for a rewritable non-volatile memory module is provided. The rewritable non-volatile memory module has a plurality of devices. The method includes dividing a plurality of memory logical units to a first logical unit group mapping to physical erasing unit of a first device and a second logical unit group mapping to physical erasing units of a second device; receiving first data stored into at least one first logical unit of the first logical unit group, and writing the first data into at least one first erasing unit, mapping to the at least one first logical unit, of the first device; and recording a wear value of every device. The method also includes performing a remapping operation corresponding to a predetermined condition if an operation status of the devices meets the predetermined condition.