Logical-to-Physical Mapping Data Structure for Sequential Writes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory sub-systems face inefficiencies in managing data accesses for sequentially written data, leading to increased resource utilization and power consumption due to individual updates and read requests in logical-to-physical mapping data structures.
Innovation Solution
Implementing a logical-to-physical mapping data structure with a 'length' field to group sequentially written data, allowing for a single physical address and length entry, reducing the need for individual updates and read requests, thereby minimizing data accesses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If individual updates and read requests are performed for each sequentially written data entry in logical-to-physical mapping data structures, then data access accuracy is maintained, but controller overhead and power consumption increase
Solution Approach 1:
The patent merges multiple individual mapping entries for sequentially written data into a single consolidated entry. Instead of maintaining separate logical-to-physical mapping entries for each sequentially written data block, the system creates one unified entry that represents the entire sequence, thereby reducing the number of individual updates and read requests the controller must process.
Solution Approach 2:
The single consolidated mapping entry serves multiple functions simultaneously: it tracks the starting physical address, stores the length of the sequential data sequence, and enables the controller to handle multiple logical addresses that map to contiguous physical locations through a single reference point, reducing overhead while maintaining access accuracy.
2Loss of time
If multiple individual read requests are issued for sequentially written data, then complete data retrieval is ensured, but access time increases
Solution Approach 1:
The system performs preliminary organization of sequentially written data into consolidated mapping entries during the write operation. By pre-grouping sequential data blocks and creating a single unified mapping entry that captures the entire sequence's starting address and length, the controller eliminates the need for multiple individual read requests during subsequent data retrieval operations.
Solution Approach 2:
The patent combines multiple read access operations into a single read request by utilizing the consolidated mapping entry. When data needs to be retrieved, the controller can access the entire sequential data range through one read operation based on the stored starting address and length, rather than issuing multiple separate read requests for each individual data block.
Data Source
AI summary
A system includes a memory device, and a processing device, operatively coupled to the memory device, to perform operations including storing, on a volatile memory device, logical-to-physical (L2P) mapping data corresponding to sequentially written data, determining whether an L2P update criterion is satisfied, and in response to determining that the L2P update criterion is satisfied, updating an L2P mapping data structure based on the L2P mapping data. The L2P mapping data structure maintains an initial logical translation unit (LTU) of the sequentially written data, and a length of the sequentially written data from the initial LTU.