Memory Control Circuit Unit Grouping Physical Erasing Units by Data Updating Frequency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional garbage collection procedures in memory storage devices do not consider data updating frequencies, leading to potential waste of system resources as regularly updated data is moved and subsequently marked as invalid.
Innovation Solution
A memory management method that groups physical erasing units based on data updating frequencies, selectively copying valid data from high-frequency units to spare units, thereby reducing data movement and improving garbage collection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional garbage collection procedure is executed without considering data updating frequency, then physical blocks are released by moving valid data, but regularly updated data may be moved and subsequently marked as invalid, resulting in waste of system resources
Solution Approach 1:
The patent applies local quality by differentiating the treatment of physical blocks based on their data updating frequency characteristics. High-frequency updated blocks are identified and excluded from garbage collection operations, while low-frequency blocks undergo normal garbage collection. This localized differentiation prevents wasting system resources on moving data that will be quickly overwritten, thereby resolving the contradiction between garbage collection efficiency and system resource waste.
Solution Approach 2:
The patent implements preliminary action by pre-classifying physical blocks into high-frequency and low-frequency categories based on their data updating patterns before executing garbage collection. This advance classification allows the system to proactively identify which blocks should be protected from garbage collection operations, preventing the waste of moving frequently updated data before it can be reused, thus improving overall system efficiency.
2Reliability
If physical blocks are frequently updated, then data freshness is maintained, but garbage collection operations cause unnecessary data movement and system resource consumption
Solution Approach 1:
The patent applies local quality by implementing different garbage collection strategies for different types of physical blocks. High-frequency updated blocks are identified through monitoring data updating patterns and are excluded from garbage collection operations, while low-frequency blocks undergo standard garbage collection. This localized approach maintains data freshness in high-frequency blocks without incurring the overhead of unnecessary data movement, thereby reducing operational complexity while preserving reliability.
Solution Approach 2:
The patent segments the population of physical blocks into distinct groups based on their data updating frequency characteristics. By dividing blocks into high-frequency and low-frequency categories, the system can apply specialized garbage collection policies to each segment. This segmentation simplifies the overall garbage collection process by avoiding the complexity of managing all blocks uniformly, while ensuring that frequently updated blocks maintain their data freshness without unnecessary intervention.
3Quantity of substance
If all valid data is moved during garbage collection, then physical blocks become available for reuse, but regularly updated data is wasted and system resources are consumed unnecessarily
Solution Approach 1:
The patent applies local quality by selectively applying garbage collection operations only to low-frequency updated physical blocks, while protecting high-frequency updated blocks from data movement. This localized approach ensures that available physical blocks are still replenished from the low-frequency group, maintaining the quantity of reusable blocks, while avoiding the waste of moving high-frequency data that would be quickly overwritten anyway, thus reducing loss of substance.
Solution Approach 2:
The patent extracts high-frequency updated physical blocks from the general garbage collection process by identifying and isolating them based on their updating patterns. This extraction allows the system to remove these specific blocks from the garbage collection candidate pool, preventing their valid data from being moved and wasted. Meanwhile, the remaining low-frequency blocks continue to undergo normal garbage collection, ensuring that sufficient physical blocks become available for reuse without the penalty of wasting frequently updated data.
Data Source
AI summary
A memory management method, a memory storage device and a memory control circuit unit are provided. The memory management method includes: grouping a plurality of non-spare physical erasing units into a first physical erasing unit and a second physical erasing unit, and a data updating frequency of the first physical erasing unit is lower than the data updating frequency of the second physical erasing unit; selecting a third physical erasing unit from the physical erasing units belonging to the first physical erasing unit; selecting a fourth physical erasing unit from spare physical erasing units, and copying valid data stored in the third physical erasing unit to the fourth physical erasing unit.