Window-Based Flash Translation Layer for SSD Memory Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large capacity non-volatile memory devices, such as solid state drives, require significant amounts of volatile memory (DRAM) to manage full logical to physical maps, which is costly and difficult to manage, especially as drive capacity increases.
Innovation Solution
Implementing a window-based flash translation layer using two separate maps in volatile memory, where the L2P front map has smaller granularity for quick access and the L2P back map has larger granularity, reducing the overall size of volatile memory needed while maintaining performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a full logical to physical (L2P) map is implemented in volatile memory, then mapping precision and access speed are improved, but the quantity of volatile memory required increases significantly
Solution Approach 1:
The patent divides the full L2P map into two separate maps: an L2P front map with smaller granularity for frequently accessed data, and an L2P back map with larger granularity for less frequently accessed data. This segmentation allows the system to maintain mapping precision for critical data while reducing the total volatile memory quantity by using coarser granularity for the back map.
Solution Approach 2:
The patent applies different granularity levels to different portions of the mapping system. The front map uses fine granularity (smaller data units) for high-performance requirements, while the back map uses coarse granularity (larger data units) for memory efficiency. This local quality differentiation resolves the contradiction by optimizing each segment for its specific purpose.
2Quantity of substance
If larger granularity is used in the L2P back map, then volatile memory size is reduced, but write performance deteriorates due to read-modify-write operations
Solution Approach 1:
By segmenting the mapping system into front and back maps, the patent isolates the performance impact of large-granularity operations to only the back map. Frequently written data can be directed to the front map with smaller granularity, maintaining write performance, while the back map benefits from reduced memory size requirements.
Solution Approach 2:
The L2P front map acts as an intermediary layer between the host and the L2P back map. It handles fine-grained write operations efficiently, preventing the need for read-modify-write operations on the back map for frequently accessed data, thus preserving write performance while still achieving memory size reduction.
Data Source
AI summary
Window based mapping is used to reduce the usage of volatile memory for storing the mapping of logical to physical addresses for accesses to data in a flash drive. Two separate mapping tables for translation of logical addresses to physical addresses (L2P), e.g., an L2P front map and an L2P back map, are used where the L2P front map acts as a window to the L2P back map. The L2P front map has smaller granularity for data accesses than the L2P back map. The data accessed using the L2P front map can allow the flash drive to function with relatively same performance as a fully mapped drive with a single mapping table.