Memory Scan Alignment Using Page-Level Last Written Page Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional memory sub-systems inaccurately determine last written page (LWP) data, leading to misaligned scans and scan errors during media and read disturb scans, which impact reliability and trigger rates.
Innovation Solution
The memory sub-system determines LWP data at a page level per die, using a data structure to identify and correct misalignment by adjusting scan locations based on sub-block candidate groups and empty pages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional memory sub-systems determine LWP data at a block level, then the scan process is simpler and faster, but scan alignment accuracy deteriorates leading to misaligned scans and scan errors
Solution Approach 1:
The patent divides the memory device into multiple dies, with each die containing multiple planes. The LWP determination process is segmented to operate independently on each die, identifying the last written page for each die separately. This segmentation enables precise scan alignment at the page level while maintaining manageable complexity through modular processing of individual dies.
Solution Approach 2:
The patent transitions from block-level LWP determination to page-level determination, adding a dimension of granularity to the scan alignment process. By operating at the page level rather than block level, the system achieves finer control over scan alignment accuracy while using data structures to organize and manage the increased data dimensionality.
2Reliability
If conventional memory sub-systems perform scans without page-level LWP data, then the scan process is faster and requires less computational resources, but reliability deteriorates due to scan errors
Solution Approach 1:
The patent performs preliminary determination of LWP data at the page level for each die before executing the scan operation. By pre-calculating and storing the accurate LWP positions in data structures, the system ensures scan reliability is maintained while the actual scan execution remains efficient, as the alignment calculations are completed in advance.
Solution Approach 2:
The patent introduces data structures as intermediary components that store and manage LWP information for each die. These intermediaries bridge the gap between the complex page-level LWP determination process and the simplified scan execution, allowing high reliability through accurate data tracking while maintaining productivity by pre-organizing scan targets.
3Reliability
If memory sub-systems use inaccurate LWP data, then the system operation is simpler, but misaligned scans occur affecting reliability and trigger rates
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously tracks and updates LWP information for each die based on write operations. This feedback loop ensures that the LWP data remains accurate and current, enabling reliable scan alignment while the automated tracking reduces the perceived complexity through systematic data maintenance.
Solution Approach 2:
The memory sub-system performs self-service by automatically determining and updating LWP data for each die without external intervention. The system uses internal data structures to track write operations and maintain accurate LWP information, enabling reliable scans while managing complexity through autonomous operation.
Data Source
AI summary
A scan trigger event is detected. Last written page (LWP) data for one or more memory devices of a plurality of memory devices is determined. A scan on one or more pages of the one or more memory devices of the plurality of memory devices is performed based on the LWP data.