Host-Storage Mapping Table Coordination for Lower Write Amplification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing garbage collection methods in non-volatile memory modules do not consider the efficiency and performance of the host system, leading to excessive write amplification and inefficient resource utilization.
Innovation Solution
A collaborative data management method between the host system and storage device, involving a preparatory host data consolidation process to optimize data consolidation operations, reducing write amplification and improving resource utilization by leveraging the host system's knowledge of data attributes and update frequencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If garbage collection is performed on the device end to release spare physical units, then the number of free physical units is maintained, but the write amplification factor increases and the device load becomes excessive
Solution Approach 1:
The host system performs preliminary data consolidation operations before the storage device executes garbage collection. The host identifies and consolidates valid data in source physical units, updating the mapping table to reflect which physical units contain valid data. This preliminary action reduces the amount of data the storage device must process during subsequent garbage collection, thereby reducing write amplification while maintaining the number of free physical units.
Solution Approach 2:
The mapping table serves as an intermediary data structure that enables coordination between the host system and storage device. The host uses the mapping table to identify source physical units containing valid data, and updates it to communicate consolidation results. This intermediary mechanism allows the host to guide the garbage collection process, reducing unnecessary data movements and write amplification while maintaining adequate spare physical units.
2Quantity of substance
If traditional garbage collection is performed in the background, then spare physical units are released for subsequent use, but the host system efficiency and performance are not optimized
Solution Approach 1:
The invention merges the garbage collection function with host system data consolidation operations. Instead of the storage device independently performing background garbage collection, the host system performs data consolidation as part of its normal data management workflow. This merging allows the host to optimize data movements based on its knowledge of data access patterns and application requirements, improving overall host system efficiency while releasing spare physical units.
Solution Approach 2:
The host system performs self-service data consolidation operations by identifying source physical units, consolidating valid data, and updating the mapping table. This self-service approach allows the host to manage its own data consolidation needs without relying solely on device-end garbage collection, thereby improving host system efficiency and performance while maintaining adequate spare physical units in the storage device.
3Device complexity
If the storage device performs all data consolidation operations, then data management is simplified, but the device load increases and service life decreases
Solution Approach 1:
The data consolidation function is segmented between the host system and storage device. The host system handles high-level data consolidation decisions, identifies source physical units, and updates the mapping table. The storage device executes specific programming commands to reduce valid data in identified source physical units. This segmentation distributes the computational and operational load, reducing device burden and extending service life while maintaining manageable data management complexity through clear division of responsibilities.
Data Source
AI summary
The invention provides a data management method and a storage device. The method is used in a host system and includes: sending a preparatory host data consolidation command to the storage device when a number of free physical units of the storage device is less than a predetermined threshold; receiving a mapping table sent by the storage device, the mapping table indicating a source physical unit for a host data consolidation process; issuing a corresponding programming command to the storage device according to an attribute of valid data in the source physical unit to reduce the valid data in the source physical unit, and thereby updating the mapping table; and transmitting the updated mapping table to the storage device for executing a device data consolidation process.