Bitline Threshold Distributions for Accurate Memory Failure Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory failure analysis methods in memory sub-systems are inadequate due to the reliance on data that may not accurately represent failure modes and can alter failure states during analysis, leading to inaccurate results and potential stress-induced changes.
Innovation Solution
Implementing failure analysis based on bitline threshold voltage distributions logged by the memory sub-system, using a trainable classifier to derive failure types from this data, and employing neural networks, decision trees, or rule-based engines for accurate failure diagnostics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing memory failure analysis methods are used, then analysis can be performed, but the data may not accurately represent failure modes and can alter failure states during analysis
Solution Approach 1:
The patent applies preliminary action by capturing and logging bitline threshold voltage distributions at the moment of failure occurrence, before any subsequent analysis operations can alter the failure state. This ensures the failure data is preserved in its original state for accurate analysis.
Solution Approach 2:
The patent creates a copy of the failure state by logging the bitline threshold voltage distributions to memory or storage at the time of failure. This copy can then be analyzed without affecting the original failure state, allowing repeated analysis while preserving the integrity of the actual failure condition.
2Productivity
If stress-induced changes occur during analysis, then analysis can proceed, but the results become inaccurate
Solution Approach 1:
The system performs preliminary capture of the bitline threshold voltage distributions immediately when failure occurs, storing this data before any stress-induced changes can occur during subsequent analysis operations. This allows fast analysis of preserved data without compromising accuracy.
Data Source
AI summary
Described are systems and methods for memory failure analysis based on bitline threshold voltage distributions. An example method of implementing a failure type prediction model includes: receiving, by a processing device, a first failure-related dataset reflecting a first bitline threshold voltage distribution associated with a first memory device; determining, based on the first failure-related dataset, a first failure type distribution for the first memory device; creating a training dataset comprising the first failure-related dataset and the first failure type distribution; and training, using the training dataset, a failure type prediction model to determine, for a second memory device, a second failure type distribution based on a second failure-related dataset comprising second bitline threshold voltage data associated with a second memory device.