Non-Volatile Memory Error Metric Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Increasing storage density in non-volatile memory devices leads to higher susceptibility to storage and read errors, and existing error correction schemes reduce storage capacity while not effectively addressing these errors.
Innovation Solution
A method for adjusting error metrics in non-volatile memory systems by performing initial writes and reads, determining error metric values, and using error adjustment characteristics to optimize subsequent writes, including adjusting parity bits, write voltage, and using padding data, while maintaining an error adjustment characteristics table to adapt to changing error conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error correction schemes are applied to memory portions with high error rates, then data reliability is improved, but storage capacity is reduced
Solution Approach 1:
The patent applies error correction dynamically and locally to only those memory portions that exhibit high error rates, rather than applying uniform error correction to all memory. The system monitors error metrics for individual memory portions and selectively enables enhanced error correction measures (such as increased parity bits or different ECC schemes) only for problematic regions, thereby maintaining high data reliability where needed while preserving storage capacity in healthy regions.
Solution Approach 2:
The system dynamically adjusts error correction parameters based on real-time error metric monitoring. Error correction characteristics are not fixed but are adapted continuously as memory portions degrade or improve. This dynamic adjustment allows the system to apply stronger error correction only when and where necessary, optimizing the balance between reliability and storage capacity utilization.
2Quantity of substance
If storage density is increased in non-volatile memory, then storage capacity is improved, but error susceptibility increases
Solution Approach 1:
The patent implements a feedback mechanism where error metrics are continuously monitored for each memory portion, and this information feeds back into the error correction decision-making process. The system uses read operations to detect errors, updates error metrics based on detected errors, and subsequently adjusts error correction parameters for future writes to the same memory portion. This closed-loop feedback enables the system to adapt to increasing error susceptibility caused by higher storage density while maintaining overall reliability.
3Reliability
If error correction parameters are adjusted for all memory portions, then data reliability is improved, but device complexity increases
Solution Approach 1:
The patent segments the memory into multiple independent memory portions and manages error correction parameters separately for each segment. Rather than treating all memory uniformly, the system maintains independent error metrics and error correction characteristics for each memory portion, allowing granular control over error correction application. This segmentation reduces overall system complexity by localizing management tasks to individual memory portions rather than requiring complex global management.
Data Source
AI summary
Systems, methods and/or devices are used to adjust error metrics for a memory portion of non-volatile memory in a storage device. In one aspect, a first write and a first read are performed on the memory portion. In accordance with results of the first read, a first error metric value for the memory portion is determined. In accordance with a determination that the first error metric value exceeds a first threshold value, an entry for the memory portion is added to a table. After the first write, when a second write to the memory portion is performed, it is determined whether the entry for the memory portion is present in the table. In accordance with a determination that the entry for the memory portion is present in the table, the second write uses a first error adjustment characteristic that is determined in accordance with the first error metric value.