3D Memory Data Randomization for Disturbance Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Three-dimensional (3-D) memory arrays face data disturbance issues due to specific patterns that can cause damage, which are not adequately addressed by existing randomization schemes designed for two-dimensional (2-D) memory systems.
Innovation Solution
Implementing a randomization method that uses a sequence of randomization keys to randomize data across vertically-stacked memory cells, with different keys used for adjacent locations in both vertical and horizontal directions, and modifying the sequence by omitting or offsetting keys to prevent repetitive patterns that can lead to data disturbance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing 2-D memory randomization schemes are used in 3-D memory arrays, then data storage capacity is maintained, but data disturbance occurs due to specific patterns that are not adequately addressed
Solution Approach 1:
The patent extends randomization from 2-D to 3-D by applying randomization keys across vertically-stacked memory cells. The randomization scheme now operates in three dimensions (horizontal, vertical, and depth), ensuring that adjacent locations in all directions use different randomization keys, thereby preventing harmful patterns specific to 3-D memory architecture.
Solution Approach 2:
The patent applies different randomization keys to different spatial locations within the 3-D memory array. Each location (or group of locations) is randomized using a specific key from the sequence, ensuring that locally adjacent cells have different randomization patterns. This localized differentiation prevents data disturbance while maintaining overall data integrity.
2Device complexity
If the same randomization key is used for adjacent memory locations, then implementation complexity is reduced, but repetitive patterns cause data disturbance in 3-D memory
Solution Approach 1:
The patent segments the randomization process by dividing the memory array into multiple groups or regions, each processed with a different randomization key from the sequence. This segmentation ensures that adjacent locations use different keys without requiring a completely complex randomization scheme, balancing simplicity with effectiveness.
Solution Approach 2:
The patent employs a periodic sequence of randomization keys that cycles through multiple keys in a predetermined pattern. This periodic application of different keys to adjacent locations prevents repetitive harmful patterns while maintaining a systematic and manageable implementation approach.
Data Source
AI summary
In a nonvolatile memory array, such as a three-dimensional array of charge-storage memory cells, data is randomized so that data of different strings along the same bit line are randomized using different keys and portions of data along neighboring word lines are randomized using different keys. Keys may be rotated so that data of a particular word line is randomized according to different keys in different strings.