SSD Mapping Cache Using Hotness Tables for Faster Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In DRAM-less SSD products, the limited cache space of the controller's SRAM restricts the storage of mapping relationship tables, leading to inefficient data reading due to the need to search both the controller and memory device for these tables, thereby reducing read speed.
Innovation Solution
A memory system and method that utilizes a hotness table to track region access frequency, identifying active regions based on this information and sending mapping information to the host, thereby optimizing data reading efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If mapping relationship tables are stored in the controller's SRAM cache, then data reading speed is improved, but the limited cache space restricts the storage capacity
Solution Approach 1:
The patent divides the mapping relationship tables into two segments: hot mapping tables (for frequently accessed data) stored in the controller's SRAM cache, and cold mapping tables (for infrequently accessed data) stored in the memory device. This segmentation allows the system to utilize the fast access speed of SRAM for critical data while maintaining the ability to store large volumes of mapping information in the memory device, thereby resolving the contradiction between reading speed and storage capacity.
2Quantity of substance
If mapping relationship tables are stored in the memory device, then storage capacity is improved, but data reading efficiency deteriorates due to extensive searches required
Solution Approach 1:
The patent implements preliminary action by proactively identifying frequently accessed regions (hot regions) and pre-loading their corresponding mapping relationship tables into the controller's SRAM cache before actual read operations occur. The controller monitors access patterns, determines hot regions based on access frequency, and prepares the mapping tables in advance. This preliminary preparation eliminates the need for extensive searches in the memory device during actual read operations, thereby improving data reading efficiency while maintaining adequate storage capacity.
3Quantity of substance
If the controller searches both controller and memory device for mapping tables, then storage capacity is maximized, but read speed is reduced
Solution Approach 1:
The patent introduces an intermediary mechanism - the hotness table and region identification logic - that acts as a mediator between the storage system and the read operation. This intermediary monitors access patterns, identifies hot regions, and determines which mapping tables should be cached in the controller's SRAM. By using this intermediary to intelligently manage the distribution of mapping tables between the controller and memory device, the system avoids the need to search both locations for every read operation, thereby maintaining both high storage capacity and fast read speed.
Data Source
AI summary
The present disclosure discloses example memory systems and operating methods thereof, and electronic systems. An example memory system includes a memory device and a controller coupled to the memory device. The controller is configured to receive a read command from a host, access a corresponding region of the memory device according to the read command, update hotness information of the corresponding region in a hotness table, determine one or more active regions based on the hotness information of regions in the hotness table, and send mapping information corresponding to the one or more active regions to the host. The hotness information includes a number of times a region is accessed by read commands from the host. The mapping information includes a mapping relationship table from logical address to physical address.