P2L Mapping Table for Memory Subsystem Data Integrity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sub-systems face data loss and performance issues during ungraceful shutdowns due to incomplete metadata writes, especially in non-byte-addressable devices where metadata overhead is limited, and frequent metadata dumps can reduce performance and affect quality of service.
Innovation Solution
The memory sub-system manages address mapping data structures by storing a physical-to-logical (P2L) table on a volatile memory device and a logical-to-physical (L2P) table on a non-volatile device, using move status and boundary indicators to ensure metadata integrity and facilitate data recovery during ungraceful shutdowns, thereby avoiding metadata dump operations and maintaining system metadata integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If metadata dumps are performed frequently to ensure data integrity, then reliability is improved, but productivity deteriorates due to performance reduction
Solution Approach 1:
The patent performs preliminary actions by maintaining a P2L mapping table in volatile memory that tracks the latest physical locations of data. This preliminary tracking structure allows the system to avoid frequent metadata dumps to non-volatile memory, as the P2L table can be quickly updated and consulted during operations, reducing the need for resource-intensive dump operations while maintaining data integrity
Solution Approach 2:
The patent creates a copy of the mapping information by maintaining the P2L mapping table as a separate structure from the traditional L2P table. This copying approach allows the system to track data locations efficiently in volatile memory without requiring frequent writes to non-volatile memory, thus improving performance while maintaining reliability
2Reliability
If metadata overhead is increased to store complete mapping information, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent segments the mapping functionality by separating the P2L mapping table from the traditional L2P table. The P2L table is specifically dedicated to tracking physical-to-logical mappings in volatile memory, while the L2P table handles logical-to-physical mappings. This segmentation allows each table to be optimized for its specific purpose, reducing the metadata overhead required for complete mapping information while maintaining reliability
Solution Approach 2:
The patent adds another dimension to the mapping structure by introducing the P2L table as a complementary structure to the traditional L2P table. This dimensional addition allows the system to track mapping information in a more efficient manner, reducing the overhead required in each individual table while maintaining complete mapping capability through the combined structure
3Productivity
If the P2L table is maintained in volatile memory for fast access, then productivity is improved, but reliability deteriorates during ungraceful shutdowns
Solution Approach 1:
The patent performs preliminary action by maintaining the P2L mapping table in volatile memory for fast access during operations. The table is proactively updated with the latest mapping information before operations complete, ensuring that even in the event of an ungraceful shutdown, the most recent mapping state is preserved in volatile memory, enabling faster recovery compared to traditional approaches
Solution Approach 2:
The patent implements feedback mechanisms where the P2L mapping table continuously tracks and updates the latest physical locations of data. This feedback loop ensures that the volatile memory maintains current mapping information, and upon ungraceful shutdown, the system can quickly recover by consulting this updated P2L table, thereby maintaining both productivity and reliability
Data Source
AI summary
A logical-to-physical (L2P) data structure and a physical-to-logical (P2L) data structure are maintained. The L2P data structure comprises table entries that map a logical address to a physical address. The P2L data structure comprises data entries that map a physical address to a logical address. The P2L data entries also comprise a data move status, a base address, and a boundary indicator. A move operation is detected, wherein the move operation indicates that data referenced by a logical address is to be moved from a source physical address to a destination physical address. Responsive to detecting the move operation, the data move status associated with the source physical address in the P2L data structure is updated.