NAND Flash Block Calibration via Selective Page Error Triage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional NAND Flash-based SSDs face inefficiencies in block calibration, where not all pages in a block benefit from calibration, leading to unnecessary calibrations and increased bit error rates over time, especially in modern 3-D TLC or QLC NAND Flash systems.
Innovation Solution
A method that detects error counts in pages of a block, selectively calibrates only those pages with high error rates, and partitions the block into page groups for iterative calibration, avoiding unnecessary calibrations and maintaining efficient memory utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If block-level calibration is performed on all pages in a block, then read error rates are reduced, but calibration overhead and time consumption increase significantly
Solution Approach 1:
The patent segments the block calibration process into two distinct phases: (1) a fast sampling phase that reads only a subset of pages to detect calibration needs, and (2) a selective calibration phase that calibrates only specific pages requiring calibration. This segmentation eliminates the need to calibrate all pages in the block, significantly reducing calibration overhead while maintaining reliability.
Solution Approach 2:
Instead of performing complete calibration on all pages in a block, the patent applies partial calibration only to the specific pages that exhibit high error rates and require calibration. This partial action approach reduces unnecessary calibration operations on pages that already have acceptable error rates, thereby reducing time consumption while maintaining sufficient read reliability.
2Reliability
If frequent block calibrations are performed to maintain low bit error rates, then memory endurance is improved, but computational resources are consumed unnecessarily
Solution Approach 1:
The patent implements a feedback mechanism where the sampling phase continuously monitors error rates in pages and dynamically determines which pages require calibration. This feedback-driven approach ensures that calibration resources are allocated only when and where needed, avoiding unnecessary computational resource consumption while maintaining low bit error rates through targeted calibration actions.
Solution Approach 2:
The sampling mechanism enables the system to self-identify pages that require calibration without external intervention or exhaustive inspection of all pages. By allowing the error detection mechanism to automatically trigger selective calibration only for affected pages, the system reduces unnecessary computational resource consumption while maintaining reliability.
3Measurement precision
If comprehensive page inspection is performed to identify calibration candidates, then calibration accuracy is improved, but processing speed decreases
Solution Approach 1:
The patent divides the page inspection process into two segments: a fast sampling phase that inspects a representative subset of pages to quickly identify potential calibration candidates, and a verification phase that confirms which sampled pages actually require calibration. This segmented approach maintains measurement precision for identifying calibration candidates while significantly improving processing speed compared to inspecting all pages.
Solution Approach 2:
The patent performs partial inspection of only a sampled subset of pages rather than comprehensively inspecting all pages in the block. This partial action approach identifies calibration candidates with sufficient accuracy to guide selective calibration, while the speed improvement from avoiding exhaustive inspection of all pages outweighs the minor loss in identification completeness.
Data Source
AI summary
A computer-implemented method, according to one embodiment, includes: detecting that an error count resulting from reading a first page in a block of storage space in memory is above a first threshold, and reading a second page in the block of storage space. The second page is one which had a highest error count of the pages in the block of storage space following a last calibration of the block of storage space. Moreover, a determination is made as to whether an error count resulting from reading the second page is above the first threshold. In response to determining that the error count resulting from reading the second page is above the first threshold, the block of storage space is calibrated. Other systems, methods, and computer program products are described in additional embodiments.