SSD Data Placement Across NAND Page Types for Faster Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage methods using NAND flash in SSDs are inefficient in handling data with different processing requirements due to varying performance across different pages, leading to suboptimal data storage and processing efficiency.
Innovation Solution
A data writing method that determines the appropriate storage area type based on the type of data to be written, allowing for targeted storage in areas with matching performance characteristics, such as SLC, MLC, TLC, or QLC areas, and selecting idle pages or converting lower-performance blocks to higher-performance ones when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is stored in different pages (LSB, CSB, MSB) with varying performance characteristics, then data storage capacity is improved, but data processing efficiency deteriorates due to inconsistent performance across pages
Solution Approach 1:
The patent applies local quality by creating different types of storage areas (first type, second type, third type) with distinct performance characteristics within the SSD. Each storage area type is optimized for specific data types: hot data goes to high-performance areas, warm data to medium-performance areas, and cold data to low-performance areas. This resolves the contradiction by ensuring that data processing efficiency is improved for each data type without sacrificing overall storage capacity.
2Device complexity
If conventional storage methods are used without data type classification, then device complexity is reduced, but data processing efficiency deteriorates due to uniform treatment of all data
Solution Approach 1:
The patent segments the storage device into multiple storage area types based on data access patterns and performance requirements. The controller classifies data into different types (hot, warm, cold) and directs each type to appropriate storage areas. This segmentation improves data processing efficiency by matching data characteristics with storage performance without requiring complex external management systems.
Solution Approach 2:
The patent changes the parameter of storage area performance characteristics by creating multiple storage area types with different speed and durability profiles. By dynamically selecting the appropriate storage area type based on data type classification, the system achieves high data processing efficiency while maintaining manageable complexity through automated classification and routing.
3Productivity
If high-performance storage areas are used for all data, then data processing efficiency is improved, but resource utilization deteriorates due to over-provisioning
Solution Approach 1:
The patent implements local quality by assigning different performance levels to different storage areas based on local data requirements. High-performance storage areas are dedicated to hot data that requires fast access, while low-performance areas handle cold data with infrequent access. This resolves the contradiction by ensuring high resource utilization efficiency while maintaining high data processing efficiency for time-sensitive operations.
Data Source
AI summary
A data writing method includes: receiving a write command, where the write command carries a type of to-be-written data; determining, based on the type of to-be-written data, a type of storage area that is in an SSD and into which the to-be-written data is written, where the SSD includes a plurality of types of storage areas; determining, based on the type of storage area, a target storage area into which the to-be-written data is written; and writing the to-be-written data into the target storage area. In embodiments of this application, data processing efficiency can be improved.