Buffer Memory Management for Mapping Table Restoration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing buffer memory management methods in memory storage devices face inefficiencies and system overload when restoring logical address-physical address mapping tables from a buffer memory to a rewritable non-volatile memory, leading to prolonged wait times and potential write command failures.
Innovation Solution
A buffer memory management method that allocates a first and second zone in the buffer memory, allowing for continuous buffer units to store and update logical address-physical address mapping tables, enabling efficient restoration to the non-volatile memory while allowing continuous write operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the updated logical address-physical address mapping tables are temporarily stored in continuous buffer units and then restored back to the rewritable non-volatile memory module, then the restoring operation can be completed, but the copying operations will lead to system overload and excessive restoring time
Solution Approach 1:
The buffer memory is divided into multiple buffer units, and the mapping tables are segmented and stored in continuous buffer units. This segmentation allows for efficient memory management and reduces the complexity of copying operations by organizing data in a structured manner that facilitates direct restoration without excessive copying.
Solution Approach 2:
The mapping tables are temporarily stored in continuous buffer units in advance before the restoration operation. This preliminary action prepares the data in an optimal format and location, enabling the restoration process to proceed directly to the rewritable non-volatile memory module without requiring additional copying operations during the actual restoration.
2Reliability
If the controller temporarily suspends receiving data of the write command during the restore operation, then the restore operation can be completed, but the prolonged wait time may cause failure to write
Solution Approach 1:
The controller continues to receive and process write commands during the restore operation without suspension. This continuity is achieved by managing the buffer memory and restoration processes in a way that does not block the write command pipeline, thereby maintaining high write operation efficiency while ensuring restore operation completion.
Solution Approach 2:
The buffer memory acts as an intermediary between the write commands and the restoration operation. By storing mapping tables in the buffer memory, the system can handle write commands through the controller while the restoration operates independently on the buffered data, allowing both operations to proceed concurrently without interference.
3Reliability
If a large amount of mapping tables are restored at once, then the buffer memory can be cleared, but the system overload and excessive restoring time occur
Solution Approach 1:
The buffer memory is segmented into multiple buffer units, and the restoration process operates on these segmented units. This segmentation allows the system to manage buffer memory effectively by processing smaller chunks of data at a time, reducing the peak system load while ensuring that the buffer memory is eventually cleared.
Solution Approach 2:
Instead of restoring all mapping tables at once, the system performs partial restoration operations on continuous buffer units. This partial action approach reduces the instantaneous system load and avoiding excessive restoring time, while still achieving the goal of clearing the buffer memory through cumulative restoration operations.
Data Source
AI summary
A buffer memory management method, a memory control circuit unit and a memory storage device are provided. The method includes allocating a first zone and a second zone in the buffer memory for temporarily storing a plurality of logical address-physical address mapping tables and performing a restore operation on the first zone. The method also includes receiving a write command, wherein a logical address-physical address table to which a logical address indicated by the write command belongs has been temporarily stored in the first zone. The method further includes copying the logical address-physical address table into the second zone, and updating the logical address-physical address table in the second zone.