Memory Sub-System L2P Table Flush Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory flush operations in memory sub-systems cause significant latency as they temporarily suspend data access requests and take substantial time to complete, affecting the efficiency of data access from host systems.
Innovation Solution
The memory sub-system performs memory flush operations on a portion of the L2P mapping data table at a time, allowing for the processing of data access requests before proceeding with additional portions, and switches between processing flush operations and data access requests until a maximum execution time or threshold is reached.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory flush operations are performed on the entire L2P mapping data table, then data integrity and consistency are ensured, but data access latency increases significantly
Solution Approach 1:
The L2P mapping data table is divided into multiple portions, and flush operations are performed on one portion at a time rather than the entire table. This segmentation allows the system to maintain data integrity for flushed portions while enabling concurrent data access requests to be processed for other portions, thereby reducing overall data access latency.
Solution Approach 2:
The system alternates between performing flush operations on different portions of the L2P mapping data table and processing data access requests. This periodic switching between flush operations and data access processing ensures that data integrity is maintained periodically while minimizing the impact on continuous data access operations.
2Stability of the object's composition
If memory flush operations suspend data access requests, then consistency of the L2P mapping table is maintained, but system productivity decreases
Solution Approach 1:
By segmenting the L2P mapping data table into multiple portions and performing flush operations on individual portions, the system maintains consistency within each flushed portion while allowing data access requests to proceed concurrently for non-flushed portions. This segmentation prevents complete suspension of data access operations.
Solution Approach 2:
The system maintains continuous data access processing by allowing data access requests to be handled concurrently with flush operations on different portions of the L2P mapping table. This continuity ensures that the system remains productive while still performing necessary flush operations to maintain consistency.
3Reliability
If flush operations complete the entire table before processing data access requests, then data consistency is ensured, but execution time exceeds maximum thresholds
Solution Approach 1:
The flush operation is segmented into multiple smaller operations on different portions of the L2P mapping data table. By processing one portion at a time and allowing data access requests to be handled concurrently for other portions, the execution time of individual flush operations is reduced below maximum thresholds while maintaining overall data consistency.
Solution Approach 2:
The system performs preliminary flush operations on portions of the L2P mapping data table before data access requests are fully processed, rather than waiting for complete table flush. This preliminary action on segments allows the system to maintain consistency for flushed portions while avoiding excessive execution times.
Data Source
AI summary
Responsive to receiving a table flush command, a first portion of an address mapping table is identified. A first flush operation with respect to a first portion of the address mapping table is performed. Responsive to receiving at least one memory access command, flush operations for a subsequent portion of the address mapping table is suspended. At least one memory access operation specified by the at least one memory access command is performed. A second flush operation with respect to the subsequent portion of the address mapping table is performed.


