Flash Memory Block Management via Global Random Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing capacity of physical blocks in flash memory systems leads to longer data movement times, reducing the performance of flash memory storage systems, especially when frequently accessing small files or invalid data, necessitating an improvement in data writing efficiency.
Innovation Solution
A block management method that groups physical blocks into data and spare areas, configures logical units, and utilizes global random physical units to temporarily store small data, reducing the need for frequent data movement and organization, thereby shortening the time required to execute write commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If physical blocks are made larger to increase storage capacity, then storage capacity is improved, but data movement time increases and performance decreases
Solution Approach 1:
The patent segments physical blocks into smaller units called physical units (e.g., pages), and groups them into logical units. This segmentation allows data to be moved in smaller chunks rather than moving entire large physical blocks, thereby reducing data movement time while maintaining total storage capacity.
Solution Approach 2:
The patent introduces a logical unit-physical unit mapping table that adds an abstraction layer between logical addresses and physical blocks. This dimensional change allows the system to manage data at the logical unit level while physically storing in larger blocks, decoupling the storage capacity from data movement granularity.
2Quantity of substance
If physical blocks are made larger to increase storage capacity, then storage capacity is improved, but system performance decreases
Solution Approach 1:
By segmenting large physical blocks into smaller physical units and organizing them into logical units, the system can perform finer-grained data operations. This segmentation enables more efficient data management and reduces the overhead of moving entire large blocks, thereby improving system performance while preserving storage capacity.
Solution Approach 2:
The logical unit-physical unit mapping table acts as an intermediary that decouples logical data access from physical storage layout. This intermediary layer allows the system to optimize data access patterns and reduce unnecessary data movement, improving overall system performance without sacrificing storage capacity.
3Ease of operation
If small files are frequently accessed, then data accessibility is improved, but data movement time increases
Solution Approach 1:
The patent segments data into small physical units that can be independently managed and moved. When small files are frequently accessed, only the relevant small physical units need to be moved rather than entire large physical blocks, significantly reducing data movement time while maintaining easy accessibility to small files.
Solution Approach 2:
The system applies different management strategies to different data regions based on access patterns. Frequently accessed small files are managed at the logical unit level with fine-grained control, while less frequently accessed data can be managed in larger units. This local quality approach optimizes data movement for frequently accessed small files without affecting overall system capacity.
4Productivity
If physical units are frequently organized to release invalid data, then storage efficiency is improved, but write command execution time increases
Solution Approach 1:
The system performs preliminary organization of physical units into logical units during initialization or idle periods, so that when write commands are executed, the data structures are already optimized. This preliminary action reduces the need for frequent reorganization during active write operations, thereby reducing write command execution time while maintaining storage efficiency.
Solution Approach 2:
The logical unit-physical unit mapping enables continuous data writing operations without requiring frequent interruptions to reorganize physical units. The mapping table allows the system to continuously write data to logical units while managing physical unit allocation in the background, maintaining both storage efficiency and fast write performance.
Data Source
AI summary
A block management method for managing a plurality of physical blocks is provided. The method includes grouping the physical blocks into a plurality of physical units, grouping a portion of the physical units into a data area and a spare area, configuring a plurality of logical units, and grouping the logical units into a plurality of logical unit groups and configuring another portion of the physical units as a plurality of global random physical units corresponding to the logical unit groups, wherein each of the global random physical units corresponds to one of the logical unit groups. The method further includes getting the physical units from the spare area as global random substitute physical units of the global random physical units. Accordingly, the method can store data in the global random physical units or the global random substitute physical units, thereby reducing the time for executing a host write command.