Memory Data Placement Around Weak Word Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory systems face increased latency and reliability issues due to weak word lines, which can lead to critical data failures, especially in latency-sensitive operations, as error recovery operations on these lines increase read latency and the risk of timeout failures.
Innovation Solution
The memory system avoids writing critical data to memory cells coupled with weak word lines by determining which data is critical and which word lines are weak, and instead writes non-critical or dummy data to these cells, reserving strong word lines for critical data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error recovery operations are performed on weak word lines, then reliability of reading data is improved, but read latency increases
Solution Approach 1:
The patent performs preliminary classification of word lines into strong and weak categories during manufacturing testing, and pre-classifies data into critical and non-critical types. This preliminary action allows the system to proactively route critical data away from weak word lines before read operations occur, eliminating the need for error recovery operations on critical data and thus avoiding latency increases while maintaining reliability
2Quantity of substance
If critical data is stored in memory cells coupled with weak word lines, then storage capacity is maximized, but latency-sensitive operations fail
Solution Approach 1:
The patent applies local quality by differentiating storage locations based on word line strength. Critical data is selectively stored only in memory cells coupled with strong word lines, while non-critical data can be stored in memory cells coupled with weak word lines. This localized quality assurance ensures that latency-sensitive operations on critical data succeed while still utilizing the full storage capacity of the memory device
3Quantity of substance
If all word lines are used for data storage, then storage density is improved, but the risk of timeout failures increases
Solution Approach 1:
The patent changes the parameter of data classification by introducing criticality as a new dimension for data management. By changing how data is categorized and routed based on its criticality parameter and the strength parameter of word lines, the system achieves high storage density while minimizing timeout failure risks through intelligent data placement
Data Source
AI summary
Methods, systems, and devices for critical data management within a memory system are described. A memory system may avoid writing critical data to weak word lines. For example, as part of a media management operation or a host write operation (among other examples), the memory system may determine which data is critical data and may determine which word lines are weak word lines, which may refer to word lines having bit error rates that satisfy a threshold. The memory system may refrain from writing critical data to memory cells coupled with weak word lines, and may instead write non-critical or dummy data to the weak word lines. The memory system may reserve the writing of critical data to memory cells coupled with non-weak word lines, which may refer to word lines having bit error rates that fail to satisfy the threshold.