Memory Array Sensing via Segmented Wear Leveling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in efficiently managing and performing sensing operations across arrays of memory cells, particularly in partitioning and applying appropriate signals based on distance from a decoder and workload, which affects wear leveling and data retention.
Innovation Solution
A controller partitions an array of memory cells into portions and applies distinct signals based on cycle counts and location, employing different wear leveling schemes to maintain data integrity and balance workload distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sensing operations are performed uniformly across all memory cell arrays, then operational simplicity is maintained, but wear leveling efficiency deteriorates due to unequal workload distribution
Solution Approach 1:
The memory system divides memory cell arrays into multiple groups based on physical characteristics such as distance from the decoder. Each group is assigned a different sensing operation pattern, allowing customized wear leveling strategies for each segment while maintaining overall system manageability.
Solution Approach 2:
Different sensing operations are applied to different groups of memory cell arrays based on their specific characteristics. Groups closer to the decoder receive different treatment than groups farther away, optimizing wear distribution locally rather than applying a uniform approach globally.
2Reliability
If sensing operations are applied frequently to all memory cells, then data retention is improved, but device lifespan deteriorates due to increased wear
Solution Approach 1:
Sensing operations are performed periodically rather than continuously, with different periods assigned to different memory cell array groups. This periodic approach maintains data retention while reducing overall wear accumulation in the device.
Solution Approach 2:
The sensing operation frequency is made dynamic and adaptive, adjusting based on the workload and wear state of each memory cell array group. This allows the system to optimize between data retention and device lifespan by varying operation intensity as conditions change.
3Device complexity
If wear leveling schemes are applied uniformly across all memory cell arrays, then implementation complexity is reduced, but workload distribution deteriorates due to unequal access patterns
Solution Approach 1:
The memory system segments memory cell arrays into groups that are managed separately with different wear leveling schemes. This segmentation allows customized workload distribution strategies for each group while keeping the overall implementation manageable through modular organization.
Solution Approach 2:
Each group of memory cell arrays receives a wear leveling scheme tailored to its specific characteristics and workload patterns. This local customization optimizes workload distribution for each group while the systematic grouping approach keeps overall implementation complexity reasonable.
Data Source
AI summary
The present disclosure includes apparatuses and methods related to sensing operations in memory. An example apparatus can perform sensing operations on an array of memory cells by applying a first signal to a first portion of the array of memory cells and a second signal to a second portion of the array of memory cells.