SSD Threshold Estimation Using Bit Flip Counts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating optimal thresholds in solid state storage systems often fail to converge on the actual optimal threshold, leading to increased bit errors, especially under certain conditions.
Innovation Solution
The proposed solution involves a process that determines bit flip counts across multiple bins by performing reads at threshold bounds, using a minimum bin or fitted curve to estimate the optimal threshold, thereby avoiding implausible bit flips and improving convergence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing threshold estimation techniques are used, then the process is simple, but the system may fail to converge on the actual optimal threshold leading to increased bit errors
Solution Approach 1:
The patent segments the threshold estimation process into multiple discrete steps: (1) performing reads at multiple different thresholds, (2) counting bit flips for each threshold, (3) determining bit flip rates, (4) selecting thresholds with bit flip rates below a threshold value, and (5) determining the optimal threshold from the selected set. This segmentation transforms a potentially unreliable single-step estimation into a systematic multi-step process that improves convergence while maintaining manageable complexity through structured procedure.
Solution Approach 2:
The patent implements feedback by using the observed bit flip counts and bit flip rates to iteratively refine the threshold estimation. The system reads at multiple thresholds, measures the actual bit flip behavior, and uses this feedback information to identify which thresholds produce acceptable bit flip rates. This feedback mechanism ensures the estimation converges on the actual optimal threshold rather than relying on theoretical assumptions alone.
2Measurement precision
If multiple reads are performed at different thresholds to estimate optimal threshold, then accuracy improves, but the number of reads and processing time increases
Solution Approach 1:
The patent applies partial action by performing reads at multiple thresholds but only processing a subset of the collected data. Specifically, after reading at multiple thresholds and counting bit flips, the system identifies only those thresholds with bit flip rates below a predetermined threshold value and uses only this filtered subset for the final optimal threshold determination. This partial processing approach maintains high estimation accuracy while reducing the computational burden and time required compared to exhaustive analysis of all read data.
Data Source
AI summary
A bit flip count is determined for each bin in a plurality of bins, including by: (1) performing a first read on a group of solid state storage cells at a first threshold that corresponds to a lower bound for a given bin and (2) performing a second read on the group of solid state storage cells at a second threshold that corresponds to an upper bound for the given bin. A minimum is determined using the bit flip counts corresponding to the plurality of bins and the minimum is used to estimate an optimal threshold.