Memory Controller Wear Leveling Using Access-Time Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile storage devices face limited write durability due to uneven distribution of write operations, which is exacerbated by the increasing demand for large digital files, leading to rapid reaching of write durability limits.
Innovation Solution
A memory controller that selects a memory block for data storage based on pattern data generated from access times, distributing wear evenly across memory blocks by identifying 'hot' and 'cold' data patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If wear leveling is used to distribute write operations evenly, then the useful life of non-volatile memory is extended, but the device complexity increases due to management tables and control mechanisms
Solution Approach 1:
The patent changes the parameter used for wear leveling from traditional write count to access time-based pattern data. By measuring how recently data was accessed and creating patterns based on access time intervals, the system achieves more accurate wear prediction without requiring complex counting mechanisms for every write operation.
Solution Approach 2:
The patent creates pattern data as a copy or representation of actual access behavior. Instead of directly tracking every write operation to determine wear, the system generates pattern data that replicates access time characteristics, which then guides block selection for wear leveling.
2Reliability
If traditional wear leveling based on write count is used, then write durability is extended, but access time performance bottlenecks are not addressed
Solution Approach 1:
The patent transitions from static wear leveling (based on fixed write counts) to dynamic wear leveling (based on variable access times). By continuously measuring access time intervals and updating pattern data dynamically, the system adapts to changing access patterns and predicts wear more accurately, addressing both durability and performance.
Solution Approach 2:
The patent implements feedback by measuring access times, comparing them against pattern data, and using the differences to guide block selection. The system continuously monitors access behavior and adjusts wear leveling decisions based on this feedback, creating a closed-loop control system that optimizes both durability and access performance.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A storage device includes a non-volatile memory and a memory controller. The memory controller selects a selection memory block to store data in response to pattern data corresponding to a logical address associated with a write request, and provides to the non-volatile memory; a physical address of the selection memory block, as well as data and a write command associated with the write request.