Non-Volatile Memory Read Voltage Calibration via Selective Intervals

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional NAND flash memory devices require frequent calibration of read voltage thresholds, especially in newer technologies like 3D TLC or QLC, which can lead to significant overhead in terms of read operations, making it impractical for devices with a large number of blocks to maintain acceptable error rates within a reasonable time frame.

Innovation Solution

Implementing a controller that determines different calibration intervals for read voltage thresholds based on the relative probabilities of read failures due to data retention effects, allowing for selective reduction in the number of reads required for calibration, and deferring calibration for pages with low bit error rates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If read voltage threshold calibration is performed frequently to maintain acceptable error rates, then reliability is improved, but productivity deteriorates due to significant overhead in read operations

Engineering Contradiction:
Improveerror rateVSAvoidthroughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the memory device into multiple blocks and divides the calibration process into block-level and device-level operations. By calibrating individual blocks separately and maintaining calibration data in a lookup table, the system can update only specific blocks that need calibration rather than performing full-device calibration, thus reducing the impact on overall throughput while maintaining reliability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary calibration of blocks during idle periods or low-utilization times, storing calibration results in advance. This allows the calibration work to be completed beforehand, so that during high-demand periods, the system can rely on pre-computed calibration data without sacrificing throughput.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If calibration is performed for all blocks to ensure acceptable error rates, then reliability is improved, but loss of time increases making it impractical for devices with large number of blocks

Engineering Contradiction:
Improveerror rateVSAvoidcalibration time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies local quality by differentiating calibration needs across different blocks. Instead of uniformly calibrating all blocks, the system identifies and calibrates only those blocks that exhibit degradation or error patterns. This selective approach reduces total calibration time while maintaining reliability where needed.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements partial calibration by updating only a subset of blocks that require it, rather than performing excessive full calibration on all blocks. The system monitors block health and applies calibration selectively to affected blocks, reducing overall calibration time while maintaining acceptable error rates.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If each wordline is separately calibrated with multiple reads per threshold, then measurement precision is improved, but productivity deteriorates due to large number of reads required

Engineering Contradiction:
Improveread voltage threshold accuracyVSAvoidcalibration throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent merges multiple calibration reads into a single comprehensive calibration operation per block. By combining the measurement of multiple thresholds and wordlines into integrated calibration routines, the system achieves the necessary measurement precision while reducing the total number of separate read operations required.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent uses copying by creating lookup tables that store calibration results. Once calibration data is obtained through precise measurements, the results are copied and stored for reuse, eliminating the need to repeatedly perform time-consuming calibration reads for the same parameters while maintaining measurement accuracy.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS20190348130A1Techniques for reducing read voltage threshold calibration in non-volatile memory
Publication Date: 2019.11.14 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US20190348130A1 patent drawing
  • US20190348130A1 patent drawing
  • US20190348130A1 patent drawing

AI summary

A non-volatile memory includes a plurality of cells each individually capable of storing multiple bits of data including bits of multiple physical pages including at least a first page and a second page. A controller of the non-volatile memory determines a first calibration interval for a first read voltage threshold defining a bit value in the first page and a different second calibration interval for a second read voltage threshold defining a bit value in the second page. The second calibration interval has a shorter duration than the first calibration interval. The controller calibrates the first and second read voltage thresholds for the plurality of memory cells in the non-volatile memory based on the determined first and second calibration intervals.