SSD Read Threshold Retuning via Zero-One Balance Counts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Solid-State Disk (SSD) controllers face challenges in managing read thresholds effectively, leading to increased latency and potential read errors due to shifts in threshold voltage distributions over time, especially in Multi-Level Cell (MLC) flash memories, where maintaining optimal read thresholds is crucial for performance and longevity.
Innovation Solution
The implementation of zero/one balance management techniques, where SSD controllers scramble data before writing it to Non-Volatile Memory (NVM), maintaining counts of zeros and ones for each read unit, and adjusting read thresholds based on these counts to restore the balance and track shifts in threshold voltage distributions, thereby optimizing read performance and extending the lifespan of SSDs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If read thresholds are kept fixed in SSD controllers, then device complexity is reduced, but read errors increase due to threshold voltage distribution shifts over time
Solution Approach 1:
The patent implements feedback by monitoring the zero-one balance of read data and using this information to dynamically adjust read thresholds. The controller counts zeros and ones in read units, detects imbalances caused by threshold voltage shifts, and adjusts thresholds accordingly to maintain optimal read accuracy throughout the device lifecycle.
Solution Approach 2:
The patent transitions from static fixed thresholds to dynamic adjustable thresholds. Read thresholds are no longer fixed but are adjusted over time based on observed data patterns and threshold voltage distribution shifts, allowing the system to adapt to aging effects and maintain reliability.
2Reliability
If dynamic threshold adjustment is implemented, then read accuracy is improved, but latency increases due to additional processing required to track and adjust thresholds
Solution Approach 1:
The patent performs preliminary actions by maintaining counts of zeros and ones during normal read operations, so that when threshold voltage shifts are detected, the adjustment can be made quickly using pre-collected data. This reduces the latency impact by having the information ready in advance rather than collecting it when adjustment is needed.
Solution Approach 2:
The patent changes the parameter being monitored from individual read errors to the aggregate zero-one balance ratio. This parameter change allows for faster detection of threshold shifts and more efficient adjustment decisions, reducing the time overhead compared to analyzing individual error cases.
3Duration of action of stationary object
If zero-one balance management is implemented, then the lifespan of SSD is extended, but device complexity increases due to additional counting and adjustment mechanisms
Solution Approach 1:
The patent implements self-service by having the SSD controller automatically monitor its own threshold voltage distribution shifts and adjust its own read thresholds without external intervention. The zero-one balance management is performed autonomously by the controller, reducing the need for complex external management systems.
Solution Approach 2:
The patent merges the threshold management function with the existing read operation infrastructure. The zero-one counting is integrated into normal read paths, and threshold adjustments are combined with existing voltage control mechanisms, avoiding the need for separate complex management systems.
Data Source
AI summary
An SSD controller maintains a zero count and a one count, and/or in some embodiments a zero/one disparity count, for each read unit read from an SLC NVM (or the lower pages of an MLC). In an event that the read unit is uncorrectable in part due to a shift in the threshold voltage distributions away from their nominal distributions, the maintained counts enable a determination of a direction and/or a magnitude to adjust a read threshold to track the threshold voltage shift and restore the read data zero/one balance. In various embodiments, the adjusted read threshold is determined in a variety of described ways (counts, percentages) that are based on a number of described factors (determined threshold voltage distributions, known stored values, past NVM operating events). Extensions of the forgoing techniques are described for MLC memories.