Flash Memory Read Scrub With Channel Tracking for Voltage Drift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nonvolatile memories, such as flash memory, face challenges in error correction due to accumulating errors over time, which are not effectively addressed by traditional memory scrubbing techniques used in volatile memories like DRAM or SRAM, requiring a specialized approach for flash memory read scrub and channel tracking.
Innovation Solution
A method and apparatus that includes a first circuit for reading data and generating statistics during error correction, and a second circuit for updating parameters and adjusting reference voltages based on these statistics, specifically designed for nonvolatile memory like NAND flash, to reduce read error rates and track channel parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional memory scrubbing techniques are used for nonvolatile memories, then errors can be detected and corrected periodically, but read error rates increase and uncorrectable errors occur due to accumulating errors over time
Solution Approach 1:
The patent implements a feedback mechanism where read scrub operations detect errors and generate statistics about bit flip patterns. These statistics are fed back to update channel parameters, which in turn adjust the reference voltages used during normal read operations. This closed-loop feedback system continuously adapts to channel conditions, preventing error accumulation by proactively compensating for drift before uncorrectable errors occur.
Solution Approach 2:
The patent dynamically changes the reference voltage parameters based on detected error patterns. By monitoring bit flip statistics from read scrubs and updating channel parameters, the system adjusts reference voltages to compensate for channel drift. This parameter adaptation allows the system to maintain optimal read margins despite accumulating errors over time, reducing read error rates without requiring more aggressive scrubbing frequencies.
2Reliability
If read scrub frequency is increased to prevent uncorrectable errors, then error correction effectiveness improves, but performance and power consumption increase
Solution Approach 1:
The patent performs preliminary error detection and channel characterization during dedicated read scrub operations, updating channel parameters in advance. This preliminary action during low-activity periods allows the system to prepare optimized reference voltages before normal read operations, preventing error accumulation without requiring frequent interruptions to productivity. The channel tracking is performed proactively rather than reactively.
Solution Approach 2:
The patent implements periodic read scrub operations at optimized intervals rather than continuous monitoring. By performing scrubs at strategically chosen periods and using the accumulated statistics to update channel parameters, the system achieves effective error prevention with minimal performance impact. The periodic nature allows batching of error correction tasks, reducing overall productivity penalties compared to continuous or more frequent scrubbing.
3Measurement precision
If channel parameters are tracked and reference voltages are adjusted, then read accuracy improves, but device complexity increases
Solution Approach 1:
The patent implements channel tracking and reference voltage adjustment using the existing ECC decoder and read scrub infrastructure. The same hardware components used for error correction are also utilized for channel parameter estimation and reference voltage optimization. This multi-functionality approach improves read accuracy without proportionally increasing device complexity, as existing resources are leveraged for multiple purposes.
Solution Approach 2:
The system performs self-characterization by analyzing its own error patterns from read scrub operations. The ECC decoder automatically generates statistics about bit flips, and the channel parameters are updated based on this self-generated data without requiring external calibration equipment or complex external control systems. This self-service approach improves read accuracy while minimizing the addition of external complexity.
Data Source
AI summary
An apparatus having a first circuit and a second circuit is disclosed. The first circuit may be configured to (i) read data from a region of a memory circuit during a read scrub of the region and (ii) generate a plurality of statistics based on (a) the data and (b) one or more bit flips performed during an error correction of the data. The memory circuit is generally configured to store the data in a nonvolatile condition. One or more reference voltages may be used to read the data. The second circuit may be configured to (i) update a plurality of parameters of the region based on the statistics and (ii) compute updated values of the reference voltages based on the parameters.