Memory Sub-System Scan Reliability Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing gap between total bytes written (TBW) and user capacity in memory sub-systems for autonomous vehicles and IoT devices poses challenges in data reliability scanning, particularly in event recorders that require high sequential write throughput and large storage of telemetric data, making it difficult to determine the extent of scanning without compromising reliability.
Innovation Solution
A memory sub-system that utilizes a media management component to perform a reduced scan operation based on known patterns and reliability characteristics of memory portions, allowing for representative sampling and varying scan rates to assess data reliability without exhausting resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a full scan operation is performed on all memory portions to ensure data reliability, then reliability is improved, but processing resources are exhausted and productivity deteriorates
Solution Approach 1:
The memory sub-system is divided into multiple memory portions with different reliability characteristics. The scan operation is segmented to perform full scans only on first memory portions with lower reliability characteristics, while performing reduced or no scans on second memory portions with higher reliability characteristics. This segmentation allows the system to maintain overall data reliability while significantly reducing the total processing resources required for scanning.
Solution Approach 2:
Different scan operations are applied to different memory portions based on their specific reliability characteristics. First memory portions receive full scan operations due to their lower reliability, while second memory portions receive reduced scan operations due to their higher reliability. This local quality approach ensures that scanning resources are allocated efficiently according to the actual needs of each memory portion, improving overall scan efficiency without compromising critical data reliability.
2Reliability
If scan operations are performed frequently to monitor data reliability in real-time, then reliability is improved, but processing resources are exhausted
Solution Approach 1:
The memory sub-system is segmented into multiple portions with different reliability characteristics, allowing differentiated scan frequencies. First memory portions with lower reliability characteristics undergo more frequent or comprehensive scanning, while second memory portions with higher reliability characteristics undergo less frequent or reduced scanning. This segmentation enables effective real-time reliability monitoring of critical data while significantly reducing overall processing resource consumption.
Solution Approach 2:
Instead of performing full scan operations on all memory portions at all times, the system applies partial scan operations (reduced scans) on second memory portions with higher reliability characteristics. This partial action approach maintains adequate reliability monitoring for critical portions while reducing excessive processing resource consumption on less critical portions, achieving an optimal balance between monitoring effectiveness and resource efficiency.
3Quantity of substance
If the memory sub-system stores large amounts of telemetric data to meet user capacity requirements, then user capacity is improved, but the gap between TBW and user capacity increases, making reliability scanning more difficult
Solution Approach 1:
The large storage capacity is segmented into multiple memory portions with different reliability characteristics. First memory portions are allocated for storing critical telemetric data requiring high reliability, while second memory portions store less critical data. This segmentation enables targeted reliability assessment on critical portions only, making it feasible to maintain data reliability even when the overall storage capacity is very large and the TBW-user capacity gap is significant.
Solution Approach 2:
Different reliability assessment strategies are applied to different memory portions based on their characteristics and the importance of stored data. Critical memory portions receive comprehensive reliability scanning, while non-critical portions receive reduced scanning. This local quality approach allows the system to effectively manage data reliability across large storage capacities without requiring exhaustive scanning of all data, thus overcoming the difficulty posed by the large TBW-user capacity gap.
Data Source
AI summary
A system includes a memory device including a plurality of groups of memory cells and a processing device that is operatively coupled to the memory device. The processing device is to receive a request to determine a reliability of the plurality of groups of memory cells. The processing device is further to perform, in response to receipt of the request, a scan operation on a sample portion of the plurality of groups of memory cells to determine a reliability of the sample portion that is representative of the reliability of the plurality of groups of memory cells.